Re: thread lib: solaris problems.

From: williamkempf_at_[hidden]
Date: 2001-10-29 15:10:54

--- In boost_at_y..., "Peter Dimov" <pdimov_at_m...> wrote:
> From: <williamkempf_at_h...>
> > > First the failed tests - I've checked and 
pthread_cond_timedwait is
> > > returning zero, even though the condition variable is never
> > signalled.  The
> > > trouble is if I've read the docs correctly 
pthread_cond_timedwait
> > can
> > > return zero any time it wants to (a spurious wakeup).  In effect
> > the client
> > > would end up polling the condition until it becomes true.
> >
> > True in theory, but in practice this shouldn't really happen.
> > Implementations should strive to eliminate all spurious wakeups.
> 
> It is my understanding that eliminating (as opposed to minimizing) 
spurious
> wakeups is rarely feasible.

Poor wording on my part.  I said "strive" hoping to indicate that not 
all spurious wakeups could be eliminated.  They are allowed for a 
reason.  However, they should be a very infrequent occurrence.

Bill Kempf