race on pipe causing read to erroneously return 0?

[Thomas Maier-Komor <thomas at maier-komor dot de>]

Is it possible that there is somewhere a race condition in the
cygwin.dll that causes the reader on a named pipe getting a return value
of 0 from read, although another process has already opened the named
pipe for writing.

In consequence, in my application I see sporadic occurrences of SIGPIPE,
when my application closes the named pipe after getting 0 from read.

Any ideas?

TIA,
Thomas


--
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/