tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

PR 56275 (pthread_cond_timedwait is broken)

I ran into problems with pthread_cond_timedwait today, which I filed
in gnats because I don't have any more time to look at it for the
moment. The short summary is that I was getting all manner of broken
behavior (e.g. spurious wakeups that appear impossible from code
examination) and my best guess is that the waiter queues are getting
crosslinked somehow.

Also the test program never completes, and I don't think it's because
it's wrong (after the initialization part, all the mutex ops are
paired and all the condvar waits have timeouts)... so either
something's causing it to deadlock or _lwp_park is not timing out when
it should.

David A. Holland

Home | Main Index | Thread Index | Old Index