Re: Threads: monitors vs. primitives

From: William Kempf (sirwillard_at_[hidden])
Date: 2000-08-11 15:34:19

--- In boost_at_[hidden], scleary_at_j... wrote:
> > . . .  The problem here, though, is 
> > correctness.  An event based implementation is subject to 
the "lost 
> > wakeup bug".  See http://www.cs.wustl.edu/~schmidt/win32-cv-
1.html 
> > for a good article on the issue.
> 
> That *is* a good article!  I stand corrected.  One question: should 
our
> definition of CV include a signal() operation?

Mr. Schmidt is probably the leading authority on these sorts of 
issues.  He's as close to an expert on concurrent programming as 
you're going to find.  His ACE library is a marvel.  Unfortunately, I 
don't think enough attention was paid to the design with ACE, which 
is why I'm so interested in what we're doing here.

As for the signal() question... Jeremy Siek and I are discussing this 
in private e-mail right now.  (I know that the discussion really 
belongs here, it just came up during discussion about an attempt at 
writing an document page for the ConditionVariable concept that I 
took an attempt at.)  I'm personally leaning towards saying yes, 
since there are some valid uses for it.  However, I know the 
potential for misuse is extremely high, so I personally could be 
persuaded not to if the discussion moves in that direction.

> > BTW, I'm not new to the concept of a monitor.
> 
> Neither am I, of course.  :)

*chuckles*  Well, your post seemed to be suggesting that we both 
were.  I just wanted to make it clear that I'm not one of those new 
to this particular concept.