This is the mail archive of the
cygwin
mailing list for the Cygwin project.
RE: Possible race in SYSV IPC (semaphores)
> I agree. I'll do that in the next couple of days.
Thank you. Please let me know when you do, so that I can
check the relevant snapshot out.
Incidentally, have you been able to confirm the dead-locking
that I reported earlier (below)? I just ran the patched cygserver
with the "-d -l 7" arguments, along with my test case, and both
dead-locked just as before.
> The test case attached from my earlier post from today, also shows
> a race (dead-lock, actually) occurring in CYGSERVER, when it is run
> with the "-d -l 7" command line (even when it is only one application,
> the test case is using it). There is an internal disconcert between
> the cygserver threads.
>
> The log ends with this, and there is no way to stop the server
> (like with "cygrunsrv -E") other than killing it from the task
> manager.
Anton Lavrentiev
Contractor NIH/NLM/NCBI