This is the mail archive of the
cygwin@cygwin.com
mailing list for the Cygwin project.
RE: pthread_cond_wait does not relock mutex on release
- From: "Robert Collins" <robert dot collins at itdomain dot com dot au>
- To: "Michael Labhard" <ince at pacifier dot com>,<cygwin at cygwin dot com>
- Date: Thu, 18 Apr 2002 16:57:13 +1000
- Subject: RE: pthread_cond_wait does not relock mutex on release
> -----Original Message-----
> From: Michael Labhard [mailto:ince@pacifier.com]
> Sent: Wednesday, April 17, 2002 10:51 PM
> To: cygwin@cygwin.com
> Subject: Re: pthread_cond_wait does not relock mutex on release
>
>
> Robert and Gerald:
>
> Both quite right. Although adding the SAFE_PRINTF made no
> difference in the
> output, checking a condition value in a loop made all the
> difference.
> Putting a printf in the loop revealed to my surprise that
> "spurious wakeups"
> were occurring thousands of times per second. I had naively
> assumed that a
> condition would stay set until signalled. Now I'm led to wonder if a
> condition variable is any performance improvement over a
> simple loop and
> short sleep. Any thoughts?
Without knowing what you are trying to achieve, I can't comment. However: in multi-threaded programs, a condition variable will almost invariably be more efficient than a loop + sleep.
As for conditions staying set, that isn't the purpose of condition variables: they are designed to allow signalling between threads, to allow efficient use of cpu resources - and to avoid that loop+sleep approach.
Rob
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/