Subj : Re: Real cause of spurius wakeups
To   : comp.programming.threads
From : Vladimir Prus
Date : Fri Apr 01 2005 11:50 am

Alexander Terekhov wrote:

> Vladimir Prus wrote:
>> 
>> Hello,
>> 
>> can somebody explain the real reason why spurious wakeups are possible.
>> That is, why call to pthread_cond_wait can return without
>> pthread_cond_broadcast/pthread_cond_signal.
> 
> Because apart from realtime scheduling,
> pthread_cond_broadcast/pthread_cond_signal can be implemented as nop.
> 
> No kidding.

Yes, I understand. But I doubt there's any real implementation that does
this way. I'm mostly interested in a practical reasons.

> See also
> 
> http://terekhov.de/DESIGN-futex-CV.cpp
> http://terekhov.de/DESIGN-futex-CV-with-async.cancelable-wait.txt

And how spurious wakeups are possible in that implementation? Unfortunately
I can't immediately see that from 10K of code.

- Volodya

.