Subj : Re: sem_getvalue  (was Re: Using hierarchical memory ...)
To   : comp.programming.threads
From : Alexander Terekhov
Date : Mon Sep 12 2005 11:37 pm


Joe Seigh wrote:
> 
> Alexander Terekhov wrote:
> > Joe Seigh wrote:
> > [...]
> >
> >>>Conflation or non-conflation aside for a moment, I'm just curious how
> >>>your lock-free sem_getvalue() for x86 would look like. Care to share?
> >>>
> >>
> >>Depends.  The documentation for that is a little strange and it's not
> >>clear if sem_getvalue has any practical use.
> >
> >
> > http://www.opengroup.org/austin/mailarchives/ag/msg08586.html
> >
> >
> 
> Depends on what you think "synchronize memory" means for posix semaphores,

I think it means much more stricter synchronization protocol than what it 
means for mutexes... due to semaphore value exposure, not mere locking as 
for mutexes.

> what is actually observable, 

Semaphore operations (lock/unlock/getvalue) are meant to be fully-fenced
and provide illusion of remote write atomicity.

> and what you mean by "remote write atomicity".

It's what makes PC (x86/WB) != TSO.

regards,
alexander.

.