[cpp-threads] Does pthread_mutex_lock() have release emantics?

[Alexander Terekhov]

On 5/6/05, Peter Dimov <pdimov at mmltd.net> wrote:
> Alexander Terekhov wrote:
> > On 5/6/05, Peter Dimov <pdimov at mmltd.net> wrote:
> >
> > [... perverse DCSI ...]
> >
> >> Can someone please explain what do we gain by not using the JMM as
> >> is (with data races left undefined)?
> >
> > Try
> >
> > http://groups.google.de/group/comp.programming.threads/msg/63fec0c2b97c76ec
> >
> > ;-)
> 
> I don't get it. :-)

It's about fully-fenced (with internal ccacq optimization) sema-lock/wait() 
(sema-unlock() remains release-only) and acquire on sema-getvalue(). 

See proposed XBD 4.10 edit. 

Compromise, so to speak.

regards,
alexander.