From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0001.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0001.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0003.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0002.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0001.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0004.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0003.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0002.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano


From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano

From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0005.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0004.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0003.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0006.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0005.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0004.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0007.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0006.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0005.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano

From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault

From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano

From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos

From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 

From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"

From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.





From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano

From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0008.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0007.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0007.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0001.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org

From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.



From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0009.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0008.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0008.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0002.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt

From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.

From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>

From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0010.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0009.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0009.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0004.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0002.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0002.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0002.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0002.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */

From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf

From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}

From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card

From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;


From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);

From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0011.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0010.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0010.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0005.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0003.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0003.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0003.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0003.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0001.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0012.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0011.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0011.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0006.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0004.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0004.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0004.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0004.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0002.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0013.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0012.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0012.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0007.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0005.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0005.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0005.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0005.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0004.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0014.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0013.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0013.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0008.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0006.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0006.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0006.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0006.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0005.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0015.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0014.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0014.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0009.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0007.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0007.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0007.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0007.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0006.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0016.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0015.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0015.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0010.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0008.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0008.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0008.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0008.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0007.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0017.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0016.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0016.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0011.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0009.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0009.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0009.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0009.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0008.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0001.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0018.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0017.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0017.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0012.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0010.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0010.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0010.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0010.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0009.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0002.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0019.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0018.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0018.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0013.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0011.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0011.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0011.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0011.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0010.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0003.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0020.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0019.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0019.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0014.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0012.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0012.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0012.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0012.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0011.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0004.html 

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0021.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0020.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0020.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0015.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0013.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0013.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0013.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0013.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0012.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0005.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0022.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0021.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0021.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0016.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0014.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0014.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0014.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0014.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0013.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0006.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0001.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0023.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0022.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0022.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0017.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0015.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0015.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0015.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0015.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0014.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0007.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0002.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org

From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org

From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org

From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 

From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0001.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org

From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0024.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0023.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0023.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0018.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0016.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0016.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0016.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0016.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0015.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0009.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0004.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0002.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0002.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0025.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0024.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0024.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0019.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0017.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0017.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0017.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0017.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0016.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0010.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0005.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0003.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0003.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0026.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0025.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0025.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0020.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0018.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0018.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0018.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0018.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0017.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0011.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0006.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0004.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0004.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0027.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0026.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0026.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0021.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0019.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0019.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0019.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0019.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0018.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0012.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0007.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0005.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0005.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0028.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0027.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0027.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0022.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0020.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0020.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0020.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0020.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0019.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0013.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0008.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0006.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0006.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0029.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0028.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0028.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0023.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0021.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0021.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0021.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0021.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0020.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0014.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0009.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0007.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0007.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0030.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0029.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0029.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0024.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0022.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0022.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0022.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0022.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0021.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0015.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0010.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0008.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0008.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0031.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0030.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0030.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0025.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0023.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0023.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0023.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0023.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0022.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0016.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0011.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0009.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0009.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0032.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0031.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0031.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0026.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0024.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0024.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0024.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0024.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0023.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0017.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0012.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0010.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0010.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0033.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0032.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0032.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0027.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0025.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0025.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0025.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0025.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0024.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0018.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0013.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0011.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0011.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0034.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0033.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0033.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0028.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0026.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0026.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0026.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0026.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0025.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0019.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0014.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0012.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0012.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0035.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0034.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0034.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0029.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0027.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0027.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0027.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0027.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0026.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0020.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0015.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0013.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0013.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0036.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0035.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0035.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0030.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0028.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0028.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0028.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0028.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0027.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0021.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0016.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0014.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0014.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0037.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0036.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0036.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0031.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0029.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0029.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0029.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0029.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0028.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0022.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0017.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0015.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0015.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0038.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0037.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0037.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0032.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0030.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0030.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0030.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0030.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0029.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0023.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0018.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0016.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0016.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0039.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0038.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0038.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0033.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0031.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0031.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0031.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0031.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0030.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0024.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0019.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0017.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0017.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0040.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0039.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0039.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0034.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0032.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0032.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0032.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0032.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0031.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0025.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0020.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0018.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0018.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0041.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0040.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0040.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0035.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0033.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0033.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0033.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0033.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0032.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0026.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0021.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0019.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0019.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0042.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0041.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0041.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0036.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0034.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0034.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0034.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0034.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0033.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0027.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0022.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0020.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0020.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0043.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0042.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0042.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0037.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0035.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0035.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0035.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0035.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0034.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0028.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0023.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0021.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0021.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0044.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0043.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0043.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0038.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0036.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0036.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0036.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0036.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0035.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0029.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0024.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0022.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0022.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0045.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0044.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0044.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0039.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0037.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0037.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0037.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0037.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0036.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0030.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0025.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0023.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0023.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0046.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0045.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0045.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0040.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0038.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0038.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0038.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0038.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0037.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0031.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0026.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0024.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0024.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0047.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0046.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0046.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0041.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0039.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0039.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0039.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0039.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0038.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0032.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0027.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0025.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0025.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0048.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0047.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0047.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0042.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0040.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0040.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0040.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0040.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0039.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0033.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0028.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0026.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0026.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0049.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0048.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0048.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0043.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0041.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0041.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0041.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0041.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0040.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0034.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0029.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0027.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0027.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0050.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0049.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0049.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0044.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0042.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0042.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0042.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0042.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0041.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0035.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0030.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0028.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0028.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0051.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0050.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0050.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0045.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0043.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0043.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0043.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0043.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0042.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0036.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0031.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0029.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0029.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0052.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0051.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0051.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0046.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0044.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0044.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0044.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0044.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0043.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0037.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0032.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0030.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0030.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0053.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0052.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0052.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0047.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0045.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0045.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0045.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0045.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0044.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0038.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0033.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0031.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0031.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0054.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0053.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0053.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0048.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0046.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0046.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0046.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0046.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0045.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0039.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0034.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0032.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0032.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0055.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0054.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0054.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0049.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0047.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0047.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0047.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0047.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0046.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0040.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0035.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0033.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0033.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0056.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0055.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0055.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0050.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0048.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0048.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0048.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0048.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0047.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0041.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0036.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0034.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0034.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0057.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0056.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0056.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0051.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0049.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0049.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0049.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0049.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0048.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0042.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0037.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0035.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0035.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0058.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0057.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0057.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0052.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0050.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0050.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0050.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0050.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0049.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0043.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0038.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0036.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0036.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0059.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0058.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0058.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0053.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0051.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0051.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0051.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0051.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0050.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0044.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0039.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0037.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0037.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0060.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0059.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0059.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0054.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0052.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0052.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0052.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0052.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0051.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0045.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0040.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0038.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0038.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0061.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0060.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0060.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0055.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0053.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0053.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0053.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0053.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0052.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0046.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0041.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0039.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0039.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0062.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0061.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0061.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0056.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0054.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0054.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0054.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0054.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0053.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0047.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0042.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0040.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0040.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0063.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0062.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0062.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0057.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0055.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0055.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0055.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0055.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0054.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0048.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0043.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0041.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0041.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0064.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0063.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0063.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0058.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0056.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0056.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0056.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0056.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0055.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0049.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0044.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0042.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0042.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0065.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0064.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0064.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0059.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0057.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0057.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0057.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0057.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0056.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0050.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0045.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0043.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0043.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0066.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0065.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0065.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0060.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0058.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0058.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0058.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0058.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0057.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0051.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0046.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0044.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0044.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0067.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0066.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0066.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0061.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0059.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0059.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0059.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0059.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0058.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0052.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0047.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0045.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0045.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0068.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0067.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0067.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0062.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0060.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0060.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0060.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0060.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0059.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0053.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0048.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0046.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0046.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0069.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0068.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0068.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0063.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0061.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0061.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0061.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0061.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0060.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0054.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0049.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0047.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0047.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0070.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0069.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0069.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0064.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0062.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0062.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0062.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0062.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0061.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0055.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0050.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0048.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0048.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0071.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0070.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0070.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0065.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0063.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0063.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0063.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0063.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0062.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0056.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0051.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0049.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0049.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0072.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0071.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0071.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0066.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0064.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0064.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0064.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0064.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0063.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0057.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0052.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0050.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0050.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0073.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0072.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0072.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0067.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0065.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0065.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0065.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0065.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0064.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0058.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0053.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0051.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0051.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0074.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0073.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0073.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0068.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0066.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0066.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0066.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0066.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0065.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0059.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0054.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0052.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0052.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0075.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0074.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0074.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0069.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0067.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0067.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0067.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0067.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0066.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0060.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0055.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0053.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0053.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0076.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0075.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0075.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0070.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0068.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0068.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0068.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0068.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0067.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0061.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0056.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0054.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0054.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0077.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0076.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0076.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0071.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0069.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0069.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0069.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0069.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0068.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0062.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0057.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0055.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0055.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0078.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0077.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0077.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0072.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0070.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0070.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0070.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0070.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0069.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0063.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0058.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0056.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0056.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0079.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0078.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0078.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0073.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0071.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0071.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0071.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0071.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0070.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0064.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0059.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0057.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0057.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0080.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0079.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0079.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0074.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0072.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0072.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0072.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0072.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0071.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0065.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0060.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0058.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0058.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0081.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0080.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0080.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0075.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0073.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0073.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0073.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0073.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0072.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0066.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0061.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0059.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0059.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac, with the SDK's I thought were necessary.
Double clicked on the X-code project for rxtx.
Selected the target as recommended.
Selected build.
And it did. With no errors and one warning about typecasting.
I checked the targets and found it was set to intel and G5.
I copied the two files so produced into the correct location of the quad G5
MacPro. The serial port worked on a G5 with rxtx.>

I will now check the settings on X-CODE and see if I can compile for the
Intel Chip with the 64 bits turned on.

Or if anyone out there has already solved this problem I would be grateful
if they would share this with me.


-- 
Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From fabioanjos at gmail.com  Tue Nov  4 04:01:31 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Tue, 4 Nov 2008 09:01:31 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
Message-ID: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>

Dear friends,

We are using RXTX in our software since last april and we have succeded in
most cases.
But, in some clients we could observe a strage behaviour.
In normal operation, the application reads data from serial port like the
example below:

29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
18CC20E32101000 (toString)

Where 20, is the block terminator.

Suddenly, the application starts to read strage characters from the same
serial port:


29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
49 0 0 0 0 24 0 20 0 -102 2 0 0 -111

The application continues to read strange chars from the serial port,
until we restart it.

We are using the RXTX 2.1.7

Here is the code used to read serial port:


/* (non-Javadoc)

* @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)

*/

// @Override

*public* *void* serialEvent(SerialPortEvent event) {



*switch* (event.getEventType()) {

*case* SerialPortEvent.*BI*:

*case* SerialPortEvent.*OE*:

*case* SerialPortEvent.*FE*:

*case* SerialPortEvent.*PE*:

*case* SerialPortEvent.*CD*:

*case* SerialPortEvent.*CTS*:

*case* SerialPortEvent.*DSR*:

*case* SerialPortEvent.*RI*:

*case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:

*break*;

*case* SerialPortEvent.*DATA_AVAILABLE*:

*try* {

String result = "";

*byte*[] buffer = *new* *byte*[10];

*int* len = 0;

*while* ((len = getPort().getInputStream().read(buffer)) > 0) {

result += *new* String(buffer, 0, len);

buffer = *new* *byte*[10];

}

processInput(result);

} *catch* (IOException e) {

logger.error("PRPRO01: " + e.getMessage());

} *catch* (Exception e) {

logger.error("PRPRO02: " + e.getMessage());

}

*break*;

}

}

Any help would be apreciated.


Atenciosamente,
F?bio Cristiano dos Anjos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081104/46628795/attachment-0082.html 

From rien.korstanje at gmail.com  Tue Nov  4 15:35:31 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Tue, 4 Nov 2008 23:35:31 +0100
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
Message-ID: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>

F?bio,

I have encountered a similair problem when using RXTX in combination with a
bad Serial-to-USB converter. If you don't have another one at hand, you
could also use the Hyper Terminal to see what the Serial Port is delivering.

If you aren't using a Serial-to-USB converter, I don't have any suggestions.

Regards,
-Rien Korstanje


On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
fabioanjos at gmail.com> wrote:

> Dear friends,
>
> We are using RXTX in our software since last april and we have succeded in
> most cases.
> But, in some clients we could observe a strage behaviour.
> In normal operation, the application reads data from serial port like the
> example below:
>
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
> 18CC20E32101000 (toString)
>
> Where 20, is the block terminator.
>
> Suddenly, the application starts to read strage characters from the same
> serial port:
>
>
> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>
> The application continues to read strange chars from the serial port,
> until we restart it.
>
> We are using the RXTX 2.1.7
>
> Here is the code used to read serial port:
>
>
> /* (non-Javadoc)
>
> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>
> */
>
> // @Override
>
> *public* *void* serialEvent(SerialPortEvent event) {
>
>
>
> *switch* (event.getEventType()) {
>
> *case* SerialPortEvent.*BI*:
>
> *case* SerialPortEvent.*OE*:
>
> *case* SerialPortEvent.*FE*:
>
> *case* SerialPortEvent.*PE*:
>
> *case* SerialPortEvent.*CD*:
>
> *case* SerialPortEvent.*CTS*:
>
> *case* SerialPortEvent.*DSR*:
>
> *case* SerialPortEvent.*RI*:
>
> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>
> *break*;
>
> *case* SerialPortEvent.*DATA_AVAILABLE*:
>
> *try* {
>
> String result =
> "";
>
> *byte*[] buffer = *new* *byte*[10];
>
> *int* len = 0;
>
> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>
> result +=
> *new* String(buffer, 0, len);
>
> buffer =
> *new* *byte*[10];
>
> }
>
> processInput(result);
>
> }
> *catch* (IOException e) {
>
> logger.error("PRPRO01: " + e.getMessage());
>
> }
> *catch* (Exception e) {
>
> logger.error("PRPRO02: " + e.getMessage());
>
> }
>
> *break*;
>
> }
>
> }
>
> Any help would be apreciated.
>
>
> Atenciosamente,
> F?bio Cristiano dos Anjos
>
> _______________________________________________
> 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/20081104/94804b6a/attachment-0081.html 

From fabioanjos at gmail.com  Wed Nov  5 04:18:00 2008
From: fabioanjos at gmail.com (=?ISO-8859-1?Q?F=E1bio_Cristiano_dos_Anjos?=)
Date: Wed, 5 Nov 2008 09:18:00 -0200
Subject: [Rxtx] Strange characters when reading from Serial port
In-Reply-To: <a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
References: <29788b250811040301l6de43422md5e1b9db6d69bd12@mail.gmail.com>
	<a3ff907a0811041435n2ca45cfbv7e731694292cf677@mail.gmail.com>
Message-ID: <29788b250811050318n524176a9w161502bc9a2a879f@mail.gmail.com>

Rien,

Tanks for the reply.
In fact, the client is using a Serial-to-USB converter. I'll check with them
to see if there is a problem with it.
Tanks again!

Atenciosamente,
F?bio Cristiano dos Anjos


On Tue, Nov 4, 2008 at 8:35 PM, Rien Korstanje <rien.korstanje at gmail.com>wrote:

> F?bio,
>
> I have encountered a similair problem when using RXTX in combination with a
> bad Serial-to-USB converter. If you don't have another one at hand, you
> could also use the Hyper Terminal to see what the Serial Port is delivering.
>
> If you aren't using a Serial-to-USB converter, I don't have any
> suggestions.
>
> Regards,
> -Rien Korstanje
>
>
>   On Tue, Nov 4, 2008 at 12:01 PM, F?bio Cristiano dos Anjos <
> fabioanjos at gmail.com> wrote:
>
>>   Dear friends,
>>
>> We are using RXTX in our software since last april and we have succeded in
>> most cases.
>> But, in some clients we could observe a strage behaviour.
>> In normal operation, the application reads data from serial port like the
>> example below:
>>
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:53 48 49 50 32 49
>> 56 67 67 50 48 69 51 50 49 48 49 48 48 48 20  (bytes)
>> 29-10-2008 14:50:02 - DEBUG: Dado recebido na porta COM3:5012
>> 18CC20E32101000 (toString)
>>
>> Where 20, is the block terminator.
>>
>> Suddenly, the application starts to read strage characters from the same
>> serial port:
>>
>>
>> 29-10-2008 14:53:37 - DEBUG: Dado recebido na porta COM3:0 112 2 0 0 0 0 0
>> 48 54 33 -32 6 0 0 48 49 54 33 -32 6 0 0 0 0 0 0 0 0 0 57 57 40 0 0 0 0 0 48
>> 0 1 0 0 0 24 0 0 0 0 0 0 0 0 69 52 0 27 0 1 0 40 48 0 0 40 0 4 0 68 56 0 0 0
>> 0 0 0 49 50 0 0 0 0 0 0 0 0 0 0 24 0 0 0 -102 2 0 48 117 54 33 -32 6 0 48 53
>> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 88 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0 0 0 0 0 0
>> 0 0 0 11 0 11 0 40 0 24 0 64 0 24 0 68 0 1 0 56 0 0 0 6 0 0 0 -102 2 0 0 0 0
>> 0 0 -36 75 12 0 0 0 -74 -12 0 -32 5 0 0 0 0 0 0 98 2 0 0 0 0 0 0 112 2 0 0 0
>> 0 0 0 -32 5 0 0 0 0 0 0 66 2 0 0 0 0 0 0 112 2 0 0 0 0 0 -122 54 33 -32 6 0
>> 0 0 117 54 33 -32 6 0 0 0 0 0 0 0 0 0 0 0 40 0 0 0 0 0 0 0 1 0 0 0 24 0 0 0
>> 0 0 0 0 0 0 0 0 27 0 1 57 69 0 0 0 40 0 4 48 68 0 0 0 0 0 0 0 0 0 0 0 0 0 52
>> 49 0 0 0 0 24 0 20 0 -102 2 0 0 -111
>>
>> The application continues to read strange chars from the serial port,
>> until we restart it.
>>
>> We are using the RXTX 2.1.7
>>
>> Here is the code used to read serial port:
>>
>>
>> /* (non-Javadoc)
>>
>> * @see gnu.io.SerialPortEventListener#serialEvent(gnu.io.SerialPortEvent)
>>
>> */
>>
>> // @Override
>>
>> *public* *void* serialEvent(SerialPortEvent event) {
>>
>>
>>
>> *switch* (event.getEventType()) {
>>
>> *case* SerialPortEvent.*BI*:
>>
>> *case* SerialPortEvent.*OE*:
>>
>> *case* SerialPortEvent.*FE*:
>>
>> *case* SerialPortEvent.*PE*:
>>
>> *case* SerialPortEvent.*CD*:
>>
>> *case* SerialPortEvent.*CTS*:
>>
>> *case* SerialPortEvent.*DSR*:
>>
>> *case* SerialPortEvent.*RI*:
>>
>> *case* SerialPortEvent.*OUTPUT_BUFFER_EMPTY*:
>>
>> *break*;
>>
>> *case* SerialPortEvent.*DATA_AVAILABLE*:
>>
>> *try* {
>>
>> String result =
>> "";
>>
>> *byte*[] buffer = *new* *byte*[10];
>>
>> *int* len = 0;
>>
>> *while* ((len = getPort().getInputStream().read(buffer)) > 0) {
>>
>> result +=
>> *new* String(buffer, 0, len);
>>
>> buffer =
>> *new* *byte*[10];
>>
>> }
>>
>> processInput(result);
>>
>> }
>> *catch* (IOException e) {
>>
>> logger.error("PRPRO01: " + e.getMessage());
>>
>> }
>> *catch* (Exception e) {
>>
>> logger.error("PRPRO02: " + e.getMessage());
>>
>> }
>>
>> *break*;
>>
>> }
>>
>> }
>>
>> Any help would be apreciated.
>>
>>
>> Atenciosamente,
>> F?bio Cristiano dos Anjos
>>
>> _______________________________________________
>> 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/20081105/c16435bf/attachment-0081.html 

From dev+lists at humph.com  Thu Nov  6 16:15:35 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:15:35 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>

Hello, this is my first post to this list, and I would like to start  
with a small contribution.

I also work on Mac OS X and reading that I should have used the CVS  
version (to be 2.1-8, although it still reports itself as 2.1-7 for  
the moment) I got it and compiled it using Xcode 3.0.

The Xcode project needs a bit of massaging, the paths to the java  
sources have to be reset, to do so I simply got rid of all the sources  
in the project Java/PB and reimported them. The SDK paths have to be  
changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9  
for PPC.

RXTX works fine with these settings (no locks problem). Tested on  
Intel so far, as my only left PPC machine is far away...

Then prompted by the 64bit post I decided to compile a quadruple  
architecture version (a very fat one!).
This is simply done by adding 64 bit support in the target setting.
This is how you do it:

select librxtxSerial in the active target popup menu. Do command- 
option-E (alternatively Project->Edit Active Target menu item). In the  
target setting windows that opens select show: All Settings.
At the top you will then find Architectures. That is set as ppc i386.  
Double click it and tick the 64 bit. Close the window, build and you  
are done.

To check that the build process did the right thing you can run the  
command "file" on the library. So, from Terminal:

$ file librxtxSerial.jnilib
librxtxSerial.jnilib: Mach-O universal binary with 4 architectures
librxtxSerial.jnilib (for architecture i386):	Mach-O dynamically  
linked shared library i386
librxtxSerial.jnilib (for architecture ppc):	Mach-O dynamically linked  
shared library ppc
librxtxSerial.jnilib (for architecture ppc64):	Mach-O 64-bit  
dynamically linked shared library ppc64
librxtxSerial.jnilib (for architecture x86_64):	Mach-O 64-bit  
dynamically linked shared library x86_64

Giuliano



From dev+lists at humph.com  Thu Nov  6 16:43:25 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Fri, 7 Nov 2008 00:43:25 +0100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
Message-ID: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>

I realise that one detail have slipped from my previous post.

On F 7 Nov, 2008, at 00:15 , Giuliano Gavazzi wrote:

[...]
> in the project Java/PB and reimported them. The SDK paths have to be
> changed as 10.2 support is gone. So I use 10.4u for Intel and 10.3.9
> for PPC.

the SDK are selected from the same setting window used to select the  
Architecture (the one called by command-option-E).

It's only 24 hours since I found out about rxtx, so my results have  
only passed a very superficial test (a small programme that writes to  
a serial usb adapter).

The functionality of the 64bit version has been verified by calling my  
programme with the -d64 option:

java -d64 myprog

that works with the 64 bit library but not with the other one. Indeed  
with the latter I get the same:

java.lang.UnsatisfiedLinkError: /Library/Java/Extensions/ 
librxtxSerial.jnilib:  thrown while loading gnu.io.RXTXCommDriver
Exception in thread "main" java.lang.UnsatisfiedLinkError: /Library/ 
Java/Extensions/librxtxSerial.jnilib:

that Noel got.

I think that I will hang around for quite a while, as my current  
project needs serial communication in Java on Mac OS and Windows...

Giuliano


From netbeans at gatworks.com  Thu Nov  6 18:45:40 2008
From: netbeans at gatworks.com (U. George)
Date: Thu, 06 Nov 2008 20:45:40 -0500
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <49139DC4.7000609@gatworks.com>

Giuliano Gavazzi wrote:
> I realise that one detail have slipped from my previous post.
> 
> 
Gee, all this appears to be new & improved. Are there any docs to 
explain jnilib?


From dev+lists at humph.com  Mon Nov 10 13:55:10 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Mon, 10 Nov 2008 21:55:10 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and
	ImageJ
In-Reply-To: <FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>
	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
Message-ID: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>

Sent this a few days ago, but just to myself... in the meantime I  
found the source, if not the cause, of the error. See at the bottom.

There's a problem with the 64 bit version. I tested it on a iMac with  
1GB of memory (might that be the problem?) and a Manhattan USB to  
serial adaptor. While without an attached serial device on the other  
end it works, when a serial device is attached it fails:

java -d64 nulltest
Default port set to '/dev/cu.usbserial'
Stable Library
=========================================
Native lib Version = RXTX-2.1-7
Java lib Version   = RXTX-2.1-7
Found port: '/dev/cu.usbserial'
Attempting to initalize port '/dev/cu.usbserial'
We opened port /dev/cu.usbserial
first thingy worked.
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Writing "DS
" to /dev/cu.usbserial
Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
device was connected at this point
Segmentation fault

I have some other clue because when the serial device is connected  
from the start, I get the error only after the first write operation  
has been performed. Actually, the write works, it is the read  
operation that causes the failure: inputStream.read.


java -d64 -version
java version "1.5.0_16"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_16- 
b06-284)
Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_16-133, mixed mode)


Further debugging leads to these lines in read_byte_array:


	struct event_info_struct *eis = ( struct event_info_struct * )
		get_java_var( env, *jobj,"eis","J" );
	
	report_time_start();
	flag = eis->eventflags[SPE_DATA_AVAILABLE];

the eis returned by get_java_var is 0x1281f9b0 that is not accessible.  
I wonder if that is a java bug, in GetIntField.


Giuliano


From netbeans at gatworks.com  Mon Nov 10 14:52:09 2008
From: netbeans at gatworks.com (U. George)
Date: Mon, 10 Nov 2008 16:52:09 -0500
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 and	ImageJ
In-Reply-To: <78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
Message-ID: <4918AD09.50609@gatworks.com>

Probably not relevant the location accessed ( looks legit for a 32bit 
reference, )  the PC would be a better help.
any debugging tools ( like gdb ) to get stack trace? any core dump ?

> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the serial  
> device was connected at this point
> Segmentation fault


From dev+lists at humph.com  Mon Nov 10 16:20:38 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 00:20:38 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	and	ImageJ
In-Reply-To: <4918AD09.50609@gatworks.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com>
	<78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com>
	<4918AD09.50609@gatworks.com>
Message-ID: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>


On M 10 Nov, 2008, at 22:52 , U. George wrote:

> Probably not relevant the location accessed ( looks legit for a  
> 32bit reference, )  the PC would be a better help.
> any debugging tools ( like gdb ) to get stack trace? any core dump ?
>
>> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
>> serial  device was connected at this point
>> Segmentation fault

I did debug with gdb and the result is at the bottom of my prev  
message. But your message made me think.. I checked some other memory  
locations and they looked different, having the lowemost bit of the  
upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
So I added 0x100000000 to the result from GetIntField and the  
programme seems to run fine now.
Submitted a bug to Apple, but I would like to know if anyone can test  
on another 64 bit platform.

Giuliano


From Howard at QtuTrains.com  Mon Nov 10 17:06:11 2008
From: Howard at QtuTrains.com (QtuTrains)
Date: Tue, 11 Nov 2008 13:06:11 +1300
Subject: [Rxtx] MAC OS X device driver
Message-ID: <58320E55-5B04-40DE-9916-2C17EEA10025@QtuTrains.com>

I have been using RXTX for a while on MAC OS X mostly, but also on  
Windows & Linux but on the MAC I get total GUI freezing occasionally,  
but only when the serial port is in use. The serial port is of course  
a USB serial adaptor. I never get similar problems on Windows or Linux.

I assume the problem to be in the device driver for the PL2303 chip -  
I seem to remember a derogatory comment on this list a while ago, but  
cannot find it in the archives.

Can anyone advise which of the many device drivers works best with  
these.

many thanks,

Howard Amos


From Martin.Oberhuber at windriver.com  Tue Nov 11 01:47:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 11 Nov 2008 09:47:35 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
Message-ID: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>


Perhaps it's a bug to try and store (void*) in an int field
to start with? - When I'm not mistaken, Java is specified
to always use 32-bit ints, so storing the (void*) in a
long or byte[] should perhaps be considered.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: rxtx-bounces at qbang.org [mailto:rxtx-bounces at qbang.org] 
> On Behalf Of Giuliano Gavazzi
> Sent: Tuesday, November 11, 2008 12:21 AM
> To: U. George
> Cc: rxtx at qbang.org
> Subject: Re: [Rxtx] 64 bit invalid access in read. Was: MAC 
> OSX10.5 INTELand ImageJ
> 
> 
> On M 10 Nov, 2008, at 22:52 , U. George wrote:
> 
> > Probably not relevant the location accessed ( looks legit for a  
> > 32bit reference, )  the PC would be a better help.
> > any debugging tools ( like gdb ) to get stack trace? any core dump ?
> >
> >> Invalid memory access of location 1284b9b8 rip=126f0ba8 << the  
> >> serial  device was connected at this point
> >> Segmentation fault
> 
> I did debug with gdb and the result is at the bottom of my prev  
> message. But your message made me think.. I checked some 
> other memory  
> locations and they looked different, having the lowemost bit of the  
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the  
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone 
> can test  
> on another 64 bit platform.
> 
> Giuliano
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx
> 


From lfarkas at lfarkas.org  Tue Nov 11 02:47:40 2008
From: lfarkas at lfarkas.org (Farkas Levente)
Date: Tue, 11 Nov 2008 10:47:40 +0100
Subject: [Rxtx] new release in the near future?
Message-ID: <491954BC.3040101@lfarkas.org>

hi,
can we wait a version 2.1-8 release in this year?
thanks.

ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
and epel-5 (for rhel/centos-5).

-- 
  Levente                               "Si vis pacem para bellum!"


From kintel at sim.no  Tue Nov 11 03:14:28 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:14:28 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5
	INTELand	ImageJ
In-Reply-To: <460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
References: <EC65CA1F-7096-4AC6-A533-632F8D276D27@humph.com>	<FD4156FB-9487-4FC4-83DC-BDCB0F47C731@humph.com><78402EBF-D25B-4C92-A2E8-2B7B958E7469@humph.com><4918AD09.50609@gatworks.com>
	<BDD9BC1F-39FA-47FB-A771-F16DC41479CF@humph.com>
	<460801A4097E3D4CA04CC64EE6485848087B6236@ism-mail03.corp.ad.wrs.com>
Message-ID: <F87084A3-C67A-40AB-AA5F-0FAE901501CE@sim.no>

Hi,

This sounds awefully like the issues I had half a year ago. I'll  
reforward my suggested patch for that issue.
Maybe one of you have the chance to verify that it (at least  
partially) improves the 64-bit situation.

~/= Marius

--
We are Elektropeople for a better living.






From kintel at sim.no  Tue Nov 11 03:15:17 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 11:15:17 +0100
Subject: [Rxtx] Fwd:  Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
Message-ID: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>



Begin forwarded message:

> From: Marius Kintel <kintel at sim.no>
> Date: June 5, 2008 4:45:46 PM GMT+02:00
> To: rxtx at qbang.org
> Subject: Re: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
>
> Hi,
>
> I just looked into this, and found a bug in rxtx causing the  
> "Invalid memory access of location" reported by Zach.
> I wrote a quick and dirty fix for this (see attachment). The fix is  
> a corresponding fix for what Trent mentions (32-bit vs. 64-bit "eis"  
> variable), but in the native code.
>
> The attached diff is against the commapi-0-0-1 branch.
> To build it in 64-bit mode on Mac OS X, you still need to patch the  
> makefile as described by Zach.
>
> ~/= Marius
>
> --
> We are Elektropeople for a better living.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SerialImp.diff
Type: application/octet-stream
Size: 722 bytes
Desc: not available
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081111/830e9b0c/attachment-0076.obj 
-------------- next part --------------
>
> _______________________________________________
> Rxtx mailing list
> Rxtx at qbang.org
> http://mailman.qbang.org/mailman/listinfo/rxtx

~/= Marius

--
We are Elektropeople for a better living.





From dev+lists at humph.com  Tue Nov 11 04:34:33 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 12:34:33 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
Message-ID: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>

ah! Thanks! How silly of me, no bug on Apple part, but a bug in RXTX.  
I could have tried in every conceivable way to retrieve the correct  
eis value, but as it was stored incorrectly to start with....
BTW, the patch is missing one thing:

from:

int get_java_var( JNIEnv *, jobject, char *, char * );


to:


size_t get_java_var( JNIEnv *, jobject, char *, char * );

and correspondingly in the source.


Giuliano


From kintel at sim.no  Tue Nov 11 05:47:19 2008
From: kintel at sim.no (Marius Kintel)
Date: Tue, 11 Nov 2008 13:47:19 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
Message-ID: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>


On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:

> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
> RXTX. I could have tried in every conceivable way to retrieve the  
> correct eis value, but as it was stored incorrectly to start with....
> BTW, the patch is missing one thing:
> [...]
>

Cool, I haven't followed up this lately, but if it is verified to  
improve matter and if it's still not in rxtx-svn, it would be nice to  
have the updated patch applied.
This is going to bite hard now that more people migrate to 64-bit.

~/= Marius

--
We are Elektropeople for a better living.






From dev+lists at humph.com  Tue Nov 11 06:07:05 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Tue, 11 Nov 2008 14:07:05 +0100
Subject: [Rxtx] Mac 10.5.2 + JDK 1.6 (64-bit) jnilib
In-Reply-To: <9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
References: <E022F827-45F3-4BEE-ADE9-C9DE4A91C427@sim.no>
	<40E11F3A-B8EC-4C9D-9B02-A8869264ABF7@sim.no>
	<D12D8C20-CF2F-4805-8F34-43E6D7270AFC@humph.com>
	<9FE0836C-77BE-488E-8CC3-9DF1CDE79566@sim.no>
Message-ID: <EB6A6436-F8EC-4F48-858D-E54870FDE477@humph.com>


On T 11 Nov, 2008, at 13:47 , Marius Kintel wrote:

>
> On Nov 11, 2008, at 12:34 PM, Giuliano Gavazzi wrote:
>
>> ah! Thanks! How silly of me, no bug on Apple part, but a bug in  
>> RXTX. I could have tried in every conceivable way to retrieve the  
>> correct eis value, but as it was stored incorrectly to start with....
>> BTW, the patch is missing one thing:
>> [...]
>>
>
> Cool, I haven't followed up this lately, but if it is verified to  
> improve matter and if it's still not in rxtx-svn, it would be nice  
> to have the updated patch applied.
> This is going to bite hard now that more people migrate to 64-bit.

I have applied your patch manually and my patch too (or else the  
returned value looses the upper 32 bits). It seems to work here. It  
would definitely be nice to have it applied to the source from which  
the current binary is built (it's difficult to compile on Windows...  
just getting now MinGW). I can supply the universal binaries for Mac  
OS (either the lock-buggy current ones or the good CVS ones) or the  
updated Xcode project.

Giuliano


From tjarvi at qbang.org  Tue Nov 11 22:07:13 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 11 Nov 2008 22:07:13 -0700 (MST)
Subject: [Rxtx] new release in the near future?
In-Reply-To: <491954BC.3040101@lfarkas.org>
References: <491954BC.3040101@lfarkas.org>
Message-ID: <Pine.LNX.4.64.0811112200150.21785@rxtx.qbang.org>

On Tue, 11 Nov 2008, Farkas Levente wrote:

> hi,
> can we wait a version 2.1-8 release in this year?
> thanks.
>
> ps. anyway i build and pushed rxtx-2.1-0.2.7r2 into fedora-9, fedora-10
> and epel-5 (for rhel/centos-5).
>
>

Hi Farkas

Yes there will be a release this year.  I'm going through remaining bugs 
at work and preparing test binaries (win32/64 mac32/64/(ppc?) linux32/64 
and (sol32?)/64) for rxtx.org.  ? means they will be there but I'm not 
really testing them.  The process is nontrivial.

Once we put those out, the releases will spin fairly quickly until we 
(those willing to dig into code) are all happy.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Tue Nov 11 22:36:37 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Wed, 12 Nov 2008 16:36:37 +1100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
 ImageJ (Giuliano Gavazzi)
In-Reply-To: <mailman.337.1226359252.2059.rxtx@qbang.org>
Message-ID: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>


 Re: 64 bit invalid access in read. Was: MAC OSX10.5 INTEL and ImageJ
(Giuliano Gavazzi)

> I did debug with gdb and the result is at the bottom of my prev
> message. But your message made me think.. I checked some other memory
> locations and they looked different, having the lowemost bit of the
> upper 32 set, while 1284b9b8 did not. Remember, I am in 64 bit.
> So I added 0x100000000 to the result from GetIntField and the
> programme seems to run fine now.
> Submitted a bug to Apple, but I would like to know if anyone can test
> on another 64 bit platform.
> 
> Giuliano

> ***********************************
Guiliano,

I would be happy to test your code. If you send it to me as a mac archive I
will be happy to try it on a PPC machine (with 7 Gb) or an intel machine
with 32 Gb.
I will also try to follow your instructions to produce my own, this will
take me a little while.
Regards

Noel Goldsmith
Air Vehicles Division
Defence Science and Technology Organisation
506 Lorimer Street Port Melbourne Vic 3207
Ph 03 96267527 Fax 03 96267089
Mobile 0428364003
Noel.goldsmith at dsto.defence.gov.au



IMPORTANT: This email remains the property of the Australian Defence Organisation and is subject to the jurisdiction of section 70 of the CRIMES ACT 1914.  If you have received this email in error, you are requested to contact the sender and delete the email.




From dev+lists at humph.com  Wed Nov 12 01:02:48 2008
From: dev+lists at humph.com (Giuliano Gavazzi)
Date: Wed, 12 Nov 2008 09:02:48 +0100
Subject: [Rxtx] 64 bit invalid access in read. Was: MAC OSX10.5 INTEL
	ImageJ (Giuliano Gavazzi)
In-Reply-To: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
References: <C540B695.22DB%Noel.Goldsmith@dsto.defence.gov.au>
Message-ID: <F746C45E-F8FF-492A-A6D1-2802361B1B9F@humph.com>


On W 12 Nov, 2008, at 06:36 , Goldsmith, Noel wrote:
> Guiliano,
>
> I would be happy to test your code. If you send it to me as a mac  
> archive I
> will be happy to try it on a PPC machine (with 7 Gb) or an intel  
> machine
> with 32 Gb.
> I will also try to follow your instructions to produce my own, this  
> will
> take me a little while.
> Regards
>
> Noel Goldsmith

Noel, as you have seen from the development of the thread the bug was  
in RXTX code. I applied the patches suggested by Marius plus another  
required change to preserve the eis.
I have just to recompile the library for 10.4 (at the moment it is  
compiled for 10.5 only...), if you require 10.3 I'll have to wait til  
I'm home at the other machine that has got the required SDK.

Giuliano



From matthew.fullerton at schepens.harvard.edu  Wed Nov 12 13:18:45 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Wed, 12 Nov 2008 15:18:45 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
Message-ID: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>

Hi,
I have been using RXTX successfully for a week or so and today tried
to move my program to another Windows PC. Both systems are XP, both
libraries are version 2.1, both JREs are 6u10.

I'm using this section of code below (taken from the example at
http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)

while (portList.hasMoreElements()) {
         portId = (CommPortIdentifier) portList.nextElement();
         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
            if (portId.getName().equals(defaultPort)) {
               System.out.println("Found port: "+defaultPort);
               portFound = true;
               // init reader thread
               //SerialInterface reader = new SerialInterface();
            }
         }

      }
      if (!portFound) {
         System.out.println("port " + defaultPort + " not found.");
      }

      // initalize serial port
      try {
         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
      } catch (PortInUseException e) {}


My port is successfully found on the non failure system (on a USB to
Serial converter - its COM6) but on the other machine I get a not
CastableException (the target port is COM5 on the failure machine. The
failed machine also has a 2nd USB to serial converter that is used for
a modem). It is this line that causes the crash

serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);


because one cannot cast an LPRPort to a SerialPort. This I understand,
but why is CommPortIdentifier.open generating an LPRPort on one
machine, and a CommPort on the other? I really can't figure out what's
going on, and can't find any mention of the error online. I'm
desperate for help...!

Thanks,
Matt


From rien.korstanje at gmail.com  Wed Nov 12 16:46:53 2008
From: rien.korstanje at gmail.com (Rien Korstanje)
Date: Thu, 13 Nov 2008 00:46:53 +0100
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <a3ff907a0811121546n2f66acc9t675291b33a7a4f0a@mail.gmail.com>

Apologies for sending again, but by accident I replied only to Matthew
rather then the list. I'm still new to this thing.

---

Hey Matthew

this bit of code here is the problem:

while (portList.hasMoreElements()) {
        portId = (CommPortIdentifier) portList.nextElement();
        //removed code
}

This will keep spinning until the last element in the list is reached, and
thus the portId will be the last one in the list. For the example you took
the code from it won't matter but for what you do it does.

The quick and ugly way to solve your problem would be to hard code the
portname when getting the comport identifiers like:

CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);

Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
all the ports, filter out the parallel port, and list the serial port names
on an interface and make your selection there.

-Rien



On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton <
matthew.fullerton at schepens.harvard.edu> wrote:

> Hi,
> I have been using RXTX successfully for a week or so and today tried
> to move my program to another Windows PC. Both systems are XP, both
> libraries are version 2.1, both JREs are 6u10.
>
> I'm using this section of code below (taken from the example at
> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>            if (portId.getName().equals(defaultPort)) {
>               System.out.println("Found port: "+defaultPort);
>               portFound = true;
>               // init reader thread
>               //SerialInterface reader = new SerialInterface();
>            }
>         }
>
>      }
>      if (!portFound) {
>         System.out.println("port " + defaultPort + " not found.");
>      }
>
>      // initalize serial port
>      try {
>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>      } catch (PortInUseException e) {}
>
>
> My port is successfully found on the non failure system (on a USB to
> Serial converter - its COM6) but on the other machine I get a not
> CastableException (the target port is COM5 on the failure machine. The
> failed machine also has a 2nd USB to serial converter that is used for
> a modem). It is this line that causes the crash
>
> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>
>
> because one cannot cast an LPRPort to a SerialPort. This I understand,
> but why is CommPortIdentifier.open generating an LPRPort on one
> machine, and a CommPort on the other? I really can't figure out what's
> going on, and can't find any mention of the error online. I'm
> desperate for help...!
>
> Thanks,
> Matt
> _______________________________________________
> 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/20081113/e5911c8e/attachment-0074.html 

From d at daniellewilde.com  Wed Nov 12 20:59:53 2008
From: d at daniellewilde.com (danielle wilde)
Date: Thu, 13 Nov 2008 14:59:53 +1100
Subject: [Rxtx] does rxtx support 230400 baud rate?
Message-ID: <689cad490811121959u76ef6791q47515439ea2233a1@mail.gmail.com>

I've been trawling sites for about an hour trying to get this info.
can anyone advise?
thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/20630dbe/attachment-0074.html 

From donoso.d at ikusi.es  Thu Nov 13 01:44:04 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 09:44:04 +0100
Subject: [Rxtx] Timeouts problems under windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435116A@srviku004.ikusi.net>

Hello, 

 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

 

In some forums I read there is a problem with timeouts under windows.

 

Is there any solution to solve the problem?

 

 

Daniel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/6f67b0a7/attachment-0074.html 

From sebastien.jean.rxtx at gmail.com  Thu Nov 13 04:21:40 2008
From: sebastien.jean.rxtx at gmail.com (=?ISO-8859-1?Q?S=E9bastien_Jean?=)
Date: Thu, 13 Nov 2008 12:21:40 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
Message-ID: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>

On Windows XP, the problem occurs.

To solve it, you just have to enable a receive timeout long enough

e.g.

SerialPort port = ....

port.enableReceiveTimeout(3600000) // wait for one hour


Once done, every call to read will be blocking...


Baz



> Hello,
>

>
I wan't to use rxtxcomm library in my Java Project to read and write data
> from serial ports in Windows Vista.
>

>
In some forums I read there is a problem with timeouts under windows.
>

>
Is there any solution to solve the problem?
>

>

>
Daniel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/646c5a6b/attachment-0074.html 

From Steffen.DETTMER at ingenico.com  Thu Nov 13 05:29:13 2008
From: Steffen.DETTMER at ingenico.com (Steffen DETTMER)
Date: Thu, 13 Nov 2008 13:29:13 +0100
Subject: [Rxtx] Re :  Timeouts problems under windows
In-Reply-To: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
References: <db6c3330811130321h60aa5987tf4a43ae8812c1c94@mail.gmail.com>
Message-ID: <20081113122913.GA4945@elberon.bln.de.ingenico.com>

* S?bastien Jean wrote on Thu, Nov 13, 2008 at 12:21 +0100:
> * Daniel wrote:
> > I wan't to use rxtxcomm library in my Java Project to read and write data
> > from serial ports in Windows Vista.
> > In some forums I read there is a problem with timeouts under windows.
> > Is there any solution to solve the problem?
>
> On Windows XP, the problem occurs.
> To solve it, you just have to enable a receive timeout long enough
> e.g.
> SerialPort port = ....
> port.enableReceiveTimeout(3600000) // wait for one hour
> Once done, every call to read will be blocking...

but then he has a /real/ problem with timeouts, namely no useable
timeouts at all, ain't?
Which problem is it? Is it

  http://mailman.qbang.org/pipermail/rxtx/2007-April/3056951.html

so maybe this

  http://mailman.qbang.org/pipermail/rxtx/2008-October/3125772.html

workaround helps?
Or is this issue solved and the workaround not needed any longer?

oki,

Steffen


















--
 
About Ingenico Throughout the world businesses rely on Ingenico for secure and expedient electronic transaction acceptance. Ingenico products leverage proven technology, established standards and unparalleled ergonomics to provide optimal reliability, versatility and usability. This comprehensive range of products is complemented by a global array of services and partnerships, enabling businesses in a number of vertical sectors to accept transactions anywhere their business takes them.
www.ingenico.com This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.


From matthew.fullerton at schepens.harvard.edu  Thu Nov 13 06:30:20 2008
From: matthew.fullerton at schepens.harvard.edu (Matthew Fullerton)
Date: Thu, 13 Nov 2008 08:30:20 -0500
Subject: [Rxtx] LPRPort created instead of CommPort?
In-Reply-To: <a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
References: <46bdfea60811121218g2a6049ebo8e7f16dcb6194bff@mail.gmail.com>
	<a3ff907a0811121530y4106c5b2j455d7f7ade4c8258@mail.gmail.com>
Message-ID: <46bdfea60811130530w63bdaa80q86175d75aacc8be9@mail.gmail.com>

Hi Rien,
Thanks, I had only spotted that if it /didn't/ find the serial port, I
could easily end up with the parallel port, but now I see how I could
end up with it even if it really is there! The computer is at another
location so I'll find out later today if it really is this simple.

Thanks for the second pair of eyes!!!

Matt

2008/11/12 Rien Korstanje <rien.korstanje at gmail.com>:
> Hey Matthew,
>
> this bit of code here is the problem:
>
> while (portList.hasMoreElements()) {
>         portId = (CommPortIdentifier) portList.nextElement();
>         //removed code
> }
>
> This will keep spinning until the last element in the list is reached, and
> thus the portId will be the last one in the list. For the example you took
> the code from it won't matter but for what you do it does.
>
> The quick and ugly way to solve your problem would be to hard code the
> portname when getting the comport identifiers like:
>
> CommPortIdentifier portId = CommPortIdentifier.getPortIdentifier("COM4");
> SerialPort serialPort = (SerialPort) portId.open("SerialReader", 2000);
>
> Though you can also use CommPortIdentifier.getPortIdentifiers() first to get
> all the ports, filter out the parallel port, and list the serial port names
> on an interface and make your selection there.
>
> -Rien
>
>
> On Wed, Nov 12, 2008 at 9:18 PM, Matthew Fullerton
> <matthew.fullerton at schepens.harvard.edu> wrote:
>>
>> Hi,
>> I have been using RXTX successfully for a week or so and today tried
>> to move my program to another Windows PC. Both systems are XP, both
>> libraries are version 2.1, both JREs are 6u10.
>>
>> I'm using this section of code below (taken from the example at
>> http://www.captain.at/howto-java-serial-port-javax-comm-rxtx.php)
>>
>> while (portList.hasMoreElements()) {
>>         portId = (CommPortIdentifier) portList.nextElement();
>>         if (portId.getPortType() == CommPortIdentifier.PORT_SERIAL) {
>>            if (portId.getName().equals(defaultPort)) {
>>               System.out.println("Found port: "+defaultPort);
>>               portFound = true;
>>               // init reader thread
>>               //SerialInterface reader = new SerialInterface();
>>            }
>>         }
>>
>>      }
>>      if (!portFound) {
>>         System.out.println("port " + defaultPort + " not found.");
>>      }
>>
>>      // initalize serial port
>>      try {
>>         serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>      } catch (PortInUseException e) {}
>>
>>
>> My port is successfully found on the non failure system (on a USB to
>> Serial converter - its COM6) but on the other machine I get a not
>> CastableException (the target port is COM5 on the failure machine. The
>> failed machine also has a 2nd USB to serial converter that is used for
>> a modem). It is this line that causes the crash
>>
>> serialPort = (SerialPort) portId.open("SimpleReadApp", 2000);
>>
>>
>> because one cannot cast an LPRPort to a SerialPort. This I understand,
>> but why is CommPortIdentifier.open generating an LPRPort on one
>> machine, and a CommPort on the other? I really can't figure out what's
>> going on, and can't find any mention of the error online. I'm
>> desperate for help...!
>>
>> Thanks,
>> Matt
>> _______________________________________________
>> Rxtx mailing list
>> Rxtx at qbang.org
>> http://mailman.qbang.org/mailman/listinfo/rxtx
>
>


From donoso.d at ikusi.es  Thu Nov 13 06:34:02 2008
From: donoso.d at ikusi.es (Donoso Gabilondo, Daniel)
Date: Thu, 13 Nov 2008 14:34:02 +0100
Subject: [Rxtx] Timeouts problems in windows
Message-ID: <435A66F076E8344FA6D3917739BD99A40435180B@srviku004.ikusi.net>


Hello, 

I wan't to use rxtxcomm library in my Java Project to read and write
data from serial ports in Windows Vista.

In some forums I read there is a problem with timeouts under windows.

Is there any solution to solve the problem?

Daniel 


From mathieu.sacrispeyre at novamems.com  Thu Nov 13 07:22:56 2008
From: mathieu.sacrispeyre at novamems.com (Mathieu Sacrispeyre)
Date: Thu, 13 Nov 2008 15:22:56 +0100
Subject: [Rxtx] Unexpected error on unplugging of device
Message-ID: <491C3840.4090400@novamems.com>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081113/46ee151b/attachment-0073.html 

From tjarvi at mathworks.com  Thu Nov 13 13:30:40 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 15:30:40 -0500
Subject: [Rxtx] Mac patch for no driver installed.
Message-ID: <200811132030.mADKUeFV021833@jarvitlinux.mathworks.com>

Only in /local/Trash/Amake/3p/derived/win32/rxtx/src: .SerialImp.c.swo
diff -ur /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c
--- /local/Trash/Amake/3p/derived/win32/rxtx/src/SerialImp.c	2008-01-17 16:33:07.000000000 -0500
+++ /local/Trash/Amake/3p/sources/rxtx/src/SerialImp.c	2008-09-02 19:08:49.000000000 -0400
@@ -29,6 +29,7 @@
 #endif /* __LCC__ */
 #include <time.h>
 #include <stdio.h>
+#include <sys/filio.h>
 #include <string.h>
 #include <stdlib.h>
 #include <errno.h>
@@ -4462,7 +4463,7 @@
     kern_return_t    kernResult;
     mach_port_t        masterPort;
     CFMutableDictionaryRef    classesToMatch;
-    if ((kernResult=IOMasterPort(NULL, &masterPort)) != KERN_SUCCESS)
+    if ((kernResult=IOMasterPort( (int) NULL, &masterPort ) ) != KERN_SUCCESS)
     {
 	printf( "IOMasterPort returned %d\n", kernResult);
 	return kernResult;
@@ -4523,9 +4524,13 @@
     io_iterator_t    theSerialIterator;
     io_object_t      theObject;
     int              numPorts = 0;/* dima it should initiated */
-    if (createSerialIterator(&theSerialIterator) != KERN_SUCCESS)
+
+    if (( createSerialIterator( &theSerialIterator ) != KERN_SUCCESS) ||
+        ( ! IOIteratorIsValid( theSerialIterator)))
     {
-        printf( "createSerialIterator failed\n" );
+	/*  This also happens when no drivers are installed */
+        report( "createSerialIterator failed\n" );
+	return(0);
     } else {
 	jclass cls; /* dima */
 	jmethodID mid; /* dima */


From tjarvi at mathworks.com  Thu Nov 13 14:52:31 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:52:31 -0500
Subject: [Rxtx] [patch] remove some dead code.
Message-ID: <200811132152.mADLqVWB023837@jarvitlinux.mathworks.com>

Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.58
diff -u -r1.9.2.58 termios.c
--- termios.c	26 Apr 2007 05:26:06 -0000	1.9.2.58
+++ termios.c	13 Nov 2008 21:51:14 -0000
@@ -1084,26 +1084,29 @@
 	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;
 
@@ -1390,10 +1393,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1451,9 +1450,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1488,24 +1484,15 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					return -1;
 			}
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &stat);
 			return( total );
 		}
@@ -1562,10 +1549,6 @@
 			report( "vmin=0\n" );
 #endif /* DEBUG_VERBOSE */
 			ret = ClearErrors( index, &Stat);
-/*
-			usleep(1000);
-			usleep(50);
-*/
 			/* we should use -1 instead of 0 for disabled timeout */
 			now = GetTickCount();
 			if ( index->ttyset->c_cc[VTIME] &&
@@ -1625,9 +1608,6 @@
 					nBytes = 0;
 					break;
 				case ERROR_MORE_DATA:
-/*
-					usleep(1000);
-*/
 					report( "ERROR_MORE_DATA\n" );
 					break;
 				case ERROR_IO_PENDING:
@@ -1664,15 +1644,9 @@
 					}
 					sprintf(message, "end nBytes=%ld] ", nBytes);
 					report( message );
-/*
-					usleep(1000);
-*/
 					report( "ERROR_IO_PENDING\n" );
 					break;
 				default:
-/*
-					usleep(1000);
-*/
 					YACK();
 					errno = EIO;
 					printf("6\n");
@@ -1681,9 +1655,6 @@
 		}
 		else
 		{
-/*
-			usleep(1000);
-*/
 			ClearErrors( index, &Stat);
 			printf("7\n");
 			return( total );
@@ -3323,9 +3294,6 @@
 		}
 	}
 end:
-/*
-	usleep(1000);
-*/
 	LEAVE( "serial_select" );
 	return( 1 );
 #ifdef asdf


From tjarvi at mathworks.com  Thu Nov 13 14:56:25 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 16:56:25 -0500
Subject: [Rxtx] Uncle George's concurrency patch.
Message-ID: <200811132156.mADLuPEx023915@jarvitlinux.mathworks.com>


Appears to work well.

Index: RXTXPort.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXPort.java,v
retrieving revision 1.27.2.75
diff -u -r1.27.2.75 RXTXPort.java
--- RXTXPort.java	18 Nov 2007 22:32:41 -0000	1.27.2.75
+++ RXTXPort.java	13 Nov 2008 21:54:56 -0000
@@ -1,7 +1,7 @@
 /*-------------------------------------------------------------------------
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
-|   Copyright 1997-2007 by Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 1997-2008 by Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -140,6 +140,7 @@
 
 	/* dont close the file while accessing the fd */
 	int IOLocked = 0;
+	Object IOLockedMutex = new Object();
 
 	/** File descriptor */
 	private int fd = 0;
@@ -1128,25 +1129,23 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			if ( fd == 0 )
-			{
-				IOLocked--;
-				throw new IOException();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
 			}
-			try
-			{
+			try {
+				waitForTheNativeCodeSilly();
+				if ( fd == 0 )
+				{
+					throw new IOException();
+				}
 				writeByte( b, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write( int )");
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*  @param b[]
@@ -1164,20 +1163,19 @@
 				return;
 			}
 			if ( fd == 0 ) throw new IOException();
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				waitForTheNativeCodeSilly();
 				writeArray( b, 0, b.length, monThreadisInterrupted );
 				if (debug_write)
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" +b.length  +")");
+			} finally {
+				synchronized(IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 			
 		}
 	/**
@@ -1208,20 +1206,20 @@
 			{
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				writeArray( send, 0, len, monThreadisInterrupted );
 				if( debug_write )
 					z.reportln( "Leaving RXTXPort:SerialOutputStream:write(" + send.length + " " + off + " " + len + " " +") "  /*+ new String(send)*/ );
+			} finally {
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			catch( IOException e )
-			{
-				IOLocked--;
-				throw e;
-			}
-			IOLocked--;
 		}
 	/**
 	*/
@@ -1237,25 +1235,27 @@
 				z.reportln( "RXTXPort:SerialOutputStream:flush() Leaving Interrupted");
 				return;
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
-			/* 
-			   this is probably good on all OS's but for now
-			   just sendEvent from java on Sol
-			*/
+			synchronized(IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
+				/* 
+				   this is probably good on all OS's but for now
+				   just sendEvent from java on Sol
+				*/
 				if ( nativeDrain( monThreadisInterrupted ) )
 					sendEvent( SerialPortEvent.OUTPUT_BUFFER_EMPTY, true );
 				if (debug)
 					z.reportln( "RXTXPort:SerialOutputStream:flush() leave");
 			}
-			catch( IOException e )
+			finally
 			{
-				IOLocked--;
-				throw e;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
-			IOLocked--;
 		}
 	}
 
@@ -1286,14 +1286,15 @@
 			{
 				z.reportln( "+++++++++ read() monThreadisInterrupted" );
 			}
-			IOLocked++;
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() L" );
-			waitForTheNativeCodeSilly();
-			if (debug_read_results)
-				z.reportln(  "RXTXPort:SerialInputStream:read() N" );
-			try
-			{
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
+			try {
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() L" );
+				waitForTheNativeCodeSilly();
+				if (debug_read_results)
+					z.reportln(  "RXTXPort:SerialInputStream:read() N" );
 				int result = readByte();
 				if (debug_read_results)
 					//z.reportln(  "RXTXPort:SerialInputStream:read() returns byte = " + result );
@@ -1302,7 +1303,9 @@
 			}				
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1327,10 +1330,12 @@
 			{
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = read( b, 0, b.length);
 				if (debug_read_results)
 					z.reportln(  "RXTXPort:SerialInputStream:read() returned " + result + " bytes" );
@@ -1338,7 +1343,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 /*
@@ -1438,10 +1445,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readArray( b, off, Minimum);
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1449,7 +1458,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 
@@ -1547,10 +1558,12 @@
 					z.reportln( "RXTXPort:SerialInputStream:read() Interrupted");
 				return(0);
 			}
-			IOLocked++;
-			waitForTheNativeCodeSilly();
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
+				waitForTheNativeCodeSilly();
 				result = readTerminatedArray( b, off, Minimum, t );
 				if (debug_read_results)
 					z.reportln( "RXTXPort:SerialInputStream:read(" + b.length + " " + off + " " + len + ") returned " + result + " bytes"  /*+ new String(b) */);
@@ -1558,7 +1571,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	/**
@@ -1573,7 +1588,9 @@
 			}
 			if ( debug_verbose )
 				z.reportln( "RXTXPort:available() called" );
-			IOLocked++;
+			synchronized (IOLockedMutex) {
+				IOLocked++;
+			}
 			try
 			{
 				int r = nativeavailable();
@@ -1584,7 +1601,9 @@
 			}
 			finally
 			{
-				IOLocked--;
+				synchronized (IOLockedMutex) {
+					IOLocked--;
+				}
 			}
 		}
 	}


From tjarvi at mathworks.com  Thu Nov 13 15:16:11 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 17:16:11 -0500
Subject: [Rxtx] enumerated port changes.
Message-ID: <200811132216.mADMGBQI024311@jarvitlinux.mathworks.com>

move bluetooth ircomm to enumerated, add CDC/ACM devices.
http://bugzilla.qbang.org/show_bug.cgi?id=51
Index: RXTXCommDriver.java
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/gnu/io/RXTXCommDriver.java,v
retrieving revision 1.16.2.59
diff -u -r1.16.2.59 RXTXCommDriver.java
--- RXTXCommDriver.java	18 Nov 2007 22:32:41 -0000	1.16.2.59
+++ RXTXCommDriver.java	13 Nov 2008 22:15:07 -0000
@@ -2,7 +2,7 @@
 |   RXTX License v 2.1 - LGPL v 2.1 + Linking Over Controlled Interface.
 |   RXTX is a native interface to serial ports in java.
 |   Copyright 1998 Kevin Hester, kevinh at acm.org
-|   Copyright 2000-2007 Trent Jarvi tjarvi at qbang.org and others who
+|   Copyright 2000-2008 Trent Jarvi tjarvi at qbang.org and others who
 |   actually wrote it.  See individual source files for more information.
 |
 |   A copy of the LGPL v 2.1 may be found at
@@ -577,6 +577,8 @@
 						"ttyS", // linux Serial Ports
 						"ttySA", // for the IPAQs
 						"ttyUSB" // for USB frobs
+						"rfcomm",       // bluetooth serial device
+						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						};
 						CandidatePortPrefixes=Temp;
 					}
@@ -593,6 +595,7 @@
 						"ttyircomm", // linux IrCommdevices (IrDA serial emu)
 						"ttycosa0c", // linux COSA/SRP synchronous serial card
 						"ttycosa1c", // linux COSA/SRP synchronous serial card
+						"ttyACM",// linux CDC ACM devices
 						"ttyC", // linux cyclades cards
 						"ttyCH",// linux Chase Research AT/PCI-Fast serial card
 						"ttyD", // linux Digiboard serial card


From tjarvi at mathworks.com  Thu Nov 13 16:08:49 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:08:49 -0500
Subject: [Rxtx] [patch] enable threshold in SerialImp.c
Message-ID: <200811132308.mADN8nGa025308@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=85

@@ -3287,8 +3288,7 @@

        ENTER( "RXTXPort:NativeEnableRecieveTimeoutThreshold" );
        if( tcgetattr( fd, &ttyset ) < 0 ) goto fail;
-       /* TESTING ttyset.c_cc[ VMIN ] = threshold; */
-       ttyset.c_cc[ VMIN ] = 0;
+       ttyset.c_cc[ VMIN ] = threshold;
        ttyset.c_cc[ VTIME ] = timeout/100;
        if( tcsetattr( fd, TCSANOW, &ttyset ) < 0 ) goto fail;



From tjarvi at mathworks.com  Thu Nov 13 16:17:37 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 18:17:37 -0500
Subject: [Rxtx] [patch] select/EINTR
Message-ID: <200811132317.mADNHbZG025497@jarvitlinux.mathworks.com>


http://bugzilla.qbang.org/show_bug.cgi?id=89
On my PowerPC platform, it seems that a signal sometimes causes "select" to
return EINTR. I have hacked the SerialImp.c file to do some retries when it occurs.

Lines:
/* FIXME HERE Trent */
#ifndef WIN32
		ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
...
Replaced by:
/* FIXME HERE Trent */
#ifndef WIN32
		do {
			ret = SELECT( fd + 1, &rset, NULL, NULL, tvP );
		} while (ret < 0 && errno==EINTR);


From tjarvi at mathworks.com  Thu Nov 13 18:44:32 2008
From: tjarvi at mathworks.com (Trent Jarvi)
Date: Thu, 13 Nov 2008 20:44:32 -0500
Subject: [Rxtx] [patch] 1.5 stopbits fix.
Message-ID: <200811140144.mAE1iWoB006786@jarvitlinux.mathworks.com>

cvs diff: Diffing .
Index: SerialImp.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/SerialImp.c,v
retrieving revision 1.46.2.197
diff -r1.46.2.197 SerialImp.c
593c593,600
<                 case CSTOPB:  stop_bits = STOPBITS_2; break;
---
>                 case CSTOPB:  
> 			if ( ttyset.c_cflag & databits == JDATABITS_5) {
> 				stop_bits = STOPBITS_1_5;
> 			}
> 			else {
> 				stop_bits = STOPBITS_2;
> 			}
> 			break;
1130a1138
> 			(*cflag) |= CSTOPB;
Index: termios.c
===================================================================
RCS file: /usr/local/cvsroot/rxtx-devel/src/termios.c,v
retrieving revision 1.9.2.59
diff -r1.9.2.59 termios.c
1817,1818c1817,1827
< 	if ( s_termios->c_cflag & CSTOPB ) dcb->StopBits = TWOSTOPBITS;
< 		else dcb->StopBits = ONESTOPBIT;
---
> 	if ( s_termios->c_cflag & CSTOPB )
> 	{
> 		if (dcb->ByteSize == 5) 
> 		{
> 			dcb->StopBits = ONE5STOPBITS;	
> 		} else dcb->StopBits = TWOSTOPBITS;
> 	}
> 	else
> 	{
> 		dcb->StopBits = ONESTOPBIT;
> 	}
2078,2079c2087,2091
< 	}
< 	if ( myDCB.StopBits == ONESTOPBIT )
---
> 	} else if ( myDCB.StopBits == ONE5STOPBITS 
> 	{
> 		s_termios->c_cflag |= CSTOPB;	
> 		s_termios->c_cflag |= CS5;
> 	} else if ( myDCB.StopBits == ONESTOPBIT )


From tjarvi at qbang.org  Sat Nov 15 18:11:42 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Sat, 15 Nov 2008 18:11:42 -0700 (MST)
Subject: [Rxtx] CVS/Next release status
Message-ID: <Pine.LNX.4.64.0811151759450.13056@rxtx.qbang.org>


Everything I have is in CVS for the functional com aspects but I'm 
investigating one or two regressions.  They may be the same thing.  After 
that I'll just add the loadlib fixes for web installs.

The problems are related to output buffer events on w32 as far as I can 
tell so far.  More to come.

I'm also working through solaris builds - trying to eliminate some lib 
deps so they work out of the box for everyone.

Once there are no regressions, we can add anything else people need in and 
just confirm it is OK.  I can run the 'tests' after that for eveyrones 
benefit.

I've also added a couple one line fixes in CVS to the patches posted here 
if you noticed the patches have minor issues.  I was trying to make sure 
the concepts got posted individually for review and made a couple mistakes 
in the code.  I encourage others to do the same if you want fixes in.

One thing I may play with is eliminating the TIOCGETLSR Linux specific 
portion with the POSIX flush thread.  I suspect this is actually causing 
problems on Linux by sending too many events.  That may be done after the 
release.

It is tempting to just rip out a bunch of 10 year old trash in the code 
but we should do a stable release first.

--
Trent Jarvi
tjarvi at qbang.org


From rxtx at qbang.org  Thu Nov 20 12:40:23 2008
From: rxtx at qbang.org (rxtx at qbang.org)
Date: Thu, 20 Nov 2008 12:40:23 -0700
Subject: [Rxtx] MedHelp id 53066677
Message-ID: <200811201940.mAKJeN20016683@rxtx.qbang.org>

An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081120/3fecc891/attachment-0067.html 

From ehjuerrens at uni-muenster.de  Tue Nov 25 10:39:54 2008
From: ehjuerrens at uni-muenster.de (=?UTF-8?Q?Eike_Hinderk_J=C3=BCrrens?=)
Date: Tue, 25 Nov 2008 18:39:54 +0100
Subject: [Rxtx] USB to serial adapter
Message-ID: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello out there,
I am looking for an usb to serial adapter that works fine with linux
and windows using rxtx/Java.
Can you recommend any to me (not to expensive)? I will buy it here in germany.
Thank you for your help

Eike

- --
Eike Hinderk J?rrens
ehjuerrens at uni-muenster.de
N51?56'28.02'' E7?36'35.72''


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: http://getfiregpg.org

iEYEARECAAYFAkksOGgACgkQ820dQiNIWcnrrACfWh2yqOTEKR4nq1PzUXqoYtvY
rEwAn1Gri5MWqFkk+G4J4qRC8W2BJs9i
=YYbw
-----END PGP SIGNATURE-----

From Martin.Oberhuber at windriver.com  Tue Nov 25 15:45:35 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Tue, 25 Nov 2008 23:45:35 +0100
Subject: [Rxtx] rxtx unittests
Message-ID: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>

Hi Trent,
 
I remember that you said at one point that you had some testsuite for
RXTX - but I cannot find anything related in the CVS repo.
 
I would like to create some unittests for my refactoring of the port
locking code, could you let me know of any framework that this should
best fit in?
 
Thanks,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081125/2f47d945/attachment-0062.html 

From tjarvi at qbang.org  Tue Nov 25 16:22:18 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:22:18 -0700 (MST)
Subject: [Rxtx] rxtx unittests
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808A24D1E@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811251619370.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Oberhuber, Martin wrote:

> Hi Trent,
>
> I remember that you said at one point that you had some testsuite for
> RXTX - but I cannot find anything related in the CVS repo.
>
> I would like to create some unittests for my refactoring of the port
> locking code, could you let me know of any framework that this should
> best fit in?
>
> Thanks,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>
Hi Martin

There are no unit tests in rxtx at this time.  At work we have a test 
suite which depends upon MATLAB and the test harness there to test.  We 
are slowly working towards the point of having a compiled application that 
will run that test suite people can use if they like.  It will be large 
and foreign to most folks.  Differentiating rxtx test failures from
product test failures is an art.

For now, it makes sense for me to run the MATLAB testsuite and deal with 
any regressions after your changes are in.  Just let me know when you hit 
a landing spot.

Feel free to add any unit tests you have to the CVS repo.  Ideally, there 
will be a complete open source test suite eventually.  If you are adding 
the code, you get to pick the framework - code speaks.

--
Trent Jarvi
tjarvi at qbang.org



From tjarvi at qbang.org  Tue Nov 25 16:30:57 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Tue, 25 Nov 2008 16:30:57 -0700 (MST)
Subject: [Rxtx] USB to serial adapter
In-Reply-To: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
References: <87f2add00811250939p5c9a00f8l6f079ef632a5b14e@mail.gmail.com>
Message-ID: <Pine.LNX.4.64.0811251623160.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Eike Hinderk J?rrens wrote:

> Hello out there,
> I am looking for an usb to serial adapter that works fine with linux
> and windows using rxtx/Java.
> Can you recommend any to me (not to expensive)? I will buy it here in germany.
> Thank you for your help

Hi Eike

The two main types we see are based upon FTDI chips and pl2303 chips. 
You can google for 'serial linux/windows XXXX problem' and make up your 
own mind.

If the vendor has been supporting Linux/Windows for ~4 years, they are 
probably OK now.  The kernel drivers have matured.  Just make sure the 
kernel driver you use isn't 4 years old :)

Most of the 'usb serial' problems I hear about now are related to 
bluetooth.

--
Trent Jarvi
tjarvi at qbang.org

From talberts at msiscales.com  Tue Nov 25 17:10:19 2008
From: talberts at msiscales.com (Tim Alberts)
Date: Tue, 25 Nov 2008 16:10:19 -0800
Subject: [Rxtx] RXTX web install
Message-ID: <492C93EB.6030008@msiscales.com>

Greetings, I'm new to the list and soon to be new to using RXTX.  I've 
been using Sun's old javax.comm 2.0 on windows for years and haven't had 
problems, but would like to update to something supported. I'm also 
converting my applications to run via web start.  I've seen the link on:

http://users.frii.com/jarvi/rxtx/download.html

regarding DocJava.com and Dr Douglas Lyon working on an experimental web 
install.  I've used this on a few windows machines and it looks like it 
works perfectly.  I'd like to do a similar project, or help Dr Douglas 
Lyon continue with what he has.  Ideally, I need a simple way to install 
on Windows/Mac/Linux clients.

So if I can ask this mailing list if anyone has updated information on 
the support for web install.  Does what Dr Douglas Lyon provide install 
for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's 
post to this list on 20081115 (subject CVS/Next release status) that 
there will be some upcoming fixes for web install.  What are the fixes?



From tjarvi at qbang.org  Thu Nov 27 12:53:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:53:48 -0700 (MST)
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> Studio).
> Changes required were:
>
>
> *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> sources under gnu/io
> *	Fix a typo (missing character) in termios.c
>
> *	Fix includes in ParallelImp.c, win32termios.h for newer
> structure
>
> *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
>
> All changes are attached as patch. I don't think that there is an impact
> on platforms other than Windows.
> I did a quick smoke test with the build output for serial only.
> Trent: should I commit these changes?
>

Please do,

I'll test the builds with the microsoft compiler too.

I think there is a missing ) in the cvs source right now.  It should be 
obvious.  A SerialImp.c change I posted here introduced it and I forgot to 
fix it.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:54:48 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:54:48 -0700 (MST)
Subject: [Rxtx] Rev up RXTX version to 2.2
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0042@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271254190.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch revs up the RXTX version to 2.2 (and gets rid of some
> trailing whitespace).
> Ok to commit?
>
> Cheers,
> --
> Martin Oberhuber, Senior Member of Technical Staff, Wind River
> Target Management Project Lead, DSDP PMC Member
> http://www.eclipse.org/dsdp/tm
>
>
>

No objection here.

--
Trent Jarvi
tjarvi at qbang.org


From tjarvi at qbang.org  Thu Nov 27 12:56:51 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 12:56:51 -0700 (MST)
Subject: [Rxtx] RXTX web install
In-Reply-To: <492C93EB.6030008@msiscales.com>
References: <492C93EB.6030008@msiscales.com>
Message-ID: <Pine.LNX.4.64.0811271254590.11259@rxtx.qbang.org>

On Tue, 25 Nov 2008, Tim Alberts wrote:

> Greetings, I'm new to the list and soon to be new to using RXTX.  I've
> been using Sun's old javax.comm 2.0 on windows for years and haven't had
> problems, but would like to update to something supported. I'm also
> converting my applications to run via web start.  I've seen the link on:
>
> http://users.frii.com/jarvi/rxtx/download.html
>
> regarding DocJava.com and Dr Douglas Lyon working on an experimental web
> install.  I've used this on a few windows machines and it looks like it
> works perfectly.  I'd like to do a similar project, or help Dr Douglas
> Lyon continue with what he has.  Ideally, I need a simple way to install
> on Windows/Mac/Linux clients.
>
> So if I can ask this mailing list if anyone has updated information on
> the support for web install.  Does what Dr Douglas Lyon provide install
> for Windows/Mac/Linux (is it open source)?  I see from Trent Jarvi's
> post to this list on 20081115 (subject CVS/Next release status) that
> there will be some upcoming fixes for web install.  What are the fixes?
>

Hi Tim,

We do have an open source contribution comming after the functionality of 
2.2 comes in.  I also need to talk to Doug about his solution.  I 
understand Doug want's to publish how he did it.  If you can wait a bit, 
I'd like to deal with the web install as a seperate item.

--
Trent Jarvi
tjarvi at qbang.org


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:06:24 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:06:24 +0100
Subject: [Rxtx] Patches for making RXTX compile on Windows natively
In-Reply-To: <Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
References: <460801A4097E3D4CA04CC64EE648584808AB0041@ism-mail03.corp.ad.wrs.com>
	<Pine.LNX.4.64.0811271251050.11259@rxtx.qbang.org>
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0048@ism-mail03.corp.ad.wrs.com>

Done.

It builds fine now with Makefile.msvc and Makefile.mingw32.
I didn't test the mingw-cross setup.

Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 

> -----Original Message-----
> From: Trent Jarvi [mailto:tjarvi at qbang.org] 
> Sent: Thursday, November 27, 2008 8:54 PM
> To: Oberhuber, Martin
> Cc: Trent Jarvi; rxtx at qbang.org
> Subject: Re: Patches for making RXTX compile on Windows natively
> 
> On Thu, 27 Nov 2008, Oberhuber, Martin wrote:
> 
> > Hi all,
> >
> > I've retrofitted RXTX to build on Windows (with mingw32, or Visual
> > Studio).
> > Changes required were:
> >
> >
> > *	Fix and simplify Makefile.mingw32, Makefile.msvc for Java
> > sources under gnu/io
> > *	Fix a typo (missing character) in termios.c
> >
> > *	Fix includes in ParallelImp.c, win32termios.h for newer
> > structure
> >
> > *	Fix instructions (INSTALL, PORTING) for latest mingw32 and msvc
> >
> > All changes are attached as patch. I don't think that there 
> is an impact
> > on platforms other than Windows.
> > I did a quick smoke test with the build output for serial only.
> > Trent: should I commit these changes?
> >
> 
> Please do,
> 
> I'll test the builds with the microsoft compiler too.
> 
> I think there is a missing ) in the cvs source right now.  It 
> should be 
> obvious.  A SerialImp.c change I posted here introduced it 
> and I forgot to 
> fix it.
> 
> --
> Trent Jarvi
> tjarvi at qbang.org
> 


From Martin.Oberhuber at windriver.com  Thu Nov 27 13:20:34 2008
From: Martin.Oberhuber at windriver.com (Oberhuber, Martin)
Date: Thu, 27 Nov 2008 21:20:34 +0100
Subject: [Rxtx] Initial Unittests for RXTX
Message-ID: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>

Hi all,
 
attached patch adds initial Unittests for RXTX.
 
For now, there's just 3 very basic testcases (all around
getPortIdentifiers()), but they are good enough to have an initial
sanity test. The tests require that junit-3.8.2 is downloaded (from 
http://www.junit.org/). The tests are located next to src/ :
 
  rxtx-devel/src/...
  rxtx-devel/tests/gnu/io/rxtx/tests/
     RXTXTestSuite
     RXTXCommDriverTest
     CommPortIdentifierTest
 
Build support is only integrated into Makefile.msvc and Makefile.mingw32
for now -- adding build support to automake / the generic UNIX Makefile
is TODO. On Windows, tests are executed with
 
   make tests
 
Ok to commit?
 
Cheers,
--
Martin Oberhuber, Senior Member of Technical Staff, Wind River
Target Management Project Lead, DSDP PMC Member
http://www.eclipse.org/dsdp/tm
 
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0060.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: rxtx_tests.patch
Type: application/octet-stream
Size: 19947 bytes
Desc: rxtx_tests.patch
Url : http://mailman.qbang.org/pipermail/rxtx/attachments/20081127/b6776a70/attachment-0060.obj 

From tjarvi at qbang.org  Thu Nov 27 13:26:47 2008
From: tjarvi at qbang.org (Trent Jarvi)
Date: Thu, 27 Nov 2008 13:26:47 -0700 (MST)
Subject: [Rxtx] Initial Unittests for RXTX
In-Reply-To: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
References: <460801A4097E3D4CA04CC64EE648584808AB0049@ism-mail03.corp.ad.wrs.com>
Message-ID: <Pine.LNX.4.64.0811271325110.11259@rxtx.qbang.org>

On Thu, 27 Nov 2008, Oberhuber, Martin wrote:

> Hi all,
>
> attached patch adds initial Unittests for RXTX.
>
> For now, there's just 3 very basic testcases (all around
> getPortIdentifiers()), but they are good enough to have an initial
> sanity test. The tests require that junit-3.8.2 is downloaded (from
> http://www.junit.org/). The tests are located next to src/ :
>
>  rxtx-devel/src/...
>  rxtx-devel/tests/gnu/io/rxtx/tests/
>     RXTXTestSuite
>     RXTXCommDriverTest
>     CommPortIdentifierTest
>
> Build support is only integrated into Makefile.msvc and Makefile.mingw32
> for now -- adding build support to automake / the generic UNIX Makefile
> is TODO. On Windows, tests are executed with
>
>   make tests
>
> Ok to commit?
>

No objections here.  As long as the patches are discrete, I'd assume the 
patches go into CVS and anyone can flag and discuss when they have time to 
review them.

--
Trent Jarvi
tjarvi at qbang.org


From Noel.Goldsmith at dsto.defence.gov.au  Sun Nov  2 22:15:31 2008
From: Noel.Goldsmith at dsto.defence.gov.au (Goldsmith, Noel)
Date: Mon, 03 Nov 2008 16:15:31 +1100
Subject: [Rxtx] MAC OSX10.5 INTEL and ImageJ
Message-ID: <C534D423.222F%Noel.Goldsmith@dsto.defence.gov.au>

Hi,
With reference to the message on Tue Oct 21 17:41:16 MDT 2008

I am sorry I didn't include the errors I got.

Ran a plug-in to use the stage.
> 
> And using ImageJ it works fine.
> But using ImageJ64 bit I get a list of complaints.
> Note that ImageJ works fine until I try to use a serial port. I have 24 Gb
> of Ram allocated to it, and it handles it.
> 
> Now I guess this is to do with something about 64 bit.
> If you could make any suggestions that would be good.
> I am happy to try stuff out and share this with the list.

I suppose one should first ask - what are the complaints ? Is it the 24gb?

Noel Replies.

Here are the errors I get when I run ImageJ64 and try to use the stage.

java.lang.UnsatisfiedLinkError:
/Library/Java/Extensions/librxtxSerial.jnilib:
    at java.lang.ClassLoader$NativeLibrary.load(Native Method)
    at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1822)
    at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1715)
    at java.lang.Runtime.loadLibrary0(Runtime.java:822)
    at java.lang.System.loadLibrary(System.java:993)
    at gnu.io.CommPortIdentifier.<clinit>(CommPortIdentifier.java:123)
    at SerialConnection.openConnection(SerialConnection.java:84)
    at MicroComm.openConnection(MicroComm.java:147)
    at Measure_XYZ.run(Measure_XYZ.java:46)
    at ij.IJ.runUserPlugIn(IJ.java:173)
    at ij.IJ.runUserPlugIn(IJ.java:184)
    at ij.IJ.runPlugIn(IJ.java:139)
    at ij.Executer.runCommand(Executer.java:104)
    at ij.Executer.run(Executer.java:58)
    at java.lang.Thread.run(Thread.java:613)

Some of the java codes I were calling are called Measure_XYZ  and MicroComm.

I am now thinking that this failure to work might be due to the fact that I
might not have compiled rxtx for 64 bits on the intel chip (I can check when
I get to the iMac with X code on it).

If anyone has I would be grateful if they would share with me their recipe.

The following detail what I did a few months ago.

<I extracted the CVS source from the net. For rxtx 1.8 I believe.
Installed X-CODE 2.4 on a G5 iMac,