On Wed, Jun 23, 2010 at 6:13 PM, Federico G. Schwindt <[email protected]>
wrote:
>  i just wrote it but for some reason is not working as expected, which is
odd
> since it's relying on pthread_cond_timedwait. maybe the tests are wrong.
>  being 2am i think i call it a day.

I agree that it looks like it should work as well as sem_wait() does.
I'll note that the sem_* implementations in libpthread and librthread
aren't actually compliant as far as sem_post() being
async-signal-safe, so if the test involves trying that, then it's no
surprise that it fails.


Philip Guenther

Reply via email to