[Rxtx] Questions about the Rewrite2010

[Steffen DETTMER]

> The javax.comm API specification requires one event monitor 
> thread per port. 

Where does it require this? 

   Do you mean "All the events received by this listener are generated
by one dedicated thread that belongs to the SerialPort object. "?

I think this is an implementation detail, isn't it? At least it is
mentioned after "The current implementation only allows one listener per
SerialPort". A bit bad that javadoc does not distinguish between API
spec and implementation documentation. Also I don't think that
TooManyListenersException MUST be thrown (I think it CAN be thrown).

> I don't like the idea of having a thread per 
> port either, but one of the rewrite's design goals is strict 
> conformance to the specification.

(but not necessarily on JNI layer)

> The 50 mS polling interval is just a placeholder. The final 
> design of that is yet to be decided.

I though no polling would be wanted; what could be the placeholder stand
for? Do you mean a different interval duration or do you mean something
completely different?

> In Windows, you have two choices for I/O: overlapped and 
> non-overlapped. 
> In overlapped I/O a thread is blocked, waiting for something 
> to happen. 
> In non-overlapped I/O a thread does not block. I chose 
> non-overlapped I/O to avoid thread blocking at the OS level - 
> instead, the thread blocking is kept in Java.

Isn't "overlapped" (asynchronous) I/O meaning that you invoke some
ReadFile(..., &overlapped, ...)
INSTEAD of performing a synchronous (blocking) function call blocking
the thread?

As far as I see, W32_Support.cpp uses CreateFile without
FILE_FLAG_OVERLAPPED and thus "If this flag is not specified, then I/O
operations are serialized, even if the calls to the read and write
functions specify an OVERLAPPED structure." [1] and "A synchronous
handle behaves such that I/O function calls using that handle are
blocked until they complete" [2]

Given that CCOMTIMEOUTS get:
"ReadIntervalTimeout ... A value of MAXDWORD, combined with zero values
for both the ReadTotalTimeoutConstant and ReadTotalTimeoutMultiplier
members, specifies that the read operation is to return immediately with
the bytes that have already been received, even if no bytes have been
received." [3]

It looks likes this implementation relies on polling, which probably
would waste much computing power.

What did I miss?

oki,

Steffen

[1] http://msdn.microsoft.com/en-us/library/aa363858(VS.85).aspx
[2]
http://msdn.microsoft.com/en-us/library/aa363858%28VS.85%29.aspx#synchro
nous_and_asynchronous_i_o_handles
[3] http://msdn.microsoft.com/en-us/library/aa363190%28VS.85%29.aspx

 
About Ingenico: Ingenico is a leading provider of payment, transaction and business solutions, with over 15 million terminals deployed in more than 125 countries. Over 3,000 employees worldwide support merchants, banks and service providers to optimize and secure their electronic payments solutions, develop their offer of services and increase their point of sales revenue. 
http://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.
 P Please consider the environment before printing this e-mail