Hello, I've attached a small program that exhibits the behaviour, running Cygwin 1.5.18-1.
I use the select() function call to determine if it is allowed to write() to a serial port. select() indicates it is allowed, and the next write blocks. The serial port has hardware handshake control enabled (RTS/CTS), and it is unblocked as soon as CTS moves to the correct state. Is this a problem with how I've configured serial ports? I've spent a lot of time on the 'Net today and don't find anything related to this specific problem, tested it on Linux FC4 and I have the behaviour I would have expected. To compile: $ gcc -Wall wblock.c -owblock $ ./wblock Then to test I used a normal 9-pin serial cable (one that might be used to connect to a modem, or another CE device, NOT null-modem). I used a pair of scissors to just short pins 8-4 (drive CTS -ve so it writes) or 8-3 (or not at all is equivalent, drive CTS +ve so the TX line is inactive). I had an RS232 analyser (bunch of LEDs attached to the serial port) to see what was happening. Thanks, Jason.
wblock.c
Description: Binary data
cygcheck.out
Description: Binary data
-- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
