Re: Draft of generation 2 thread class

From: terekhov (terekhov_at_[hidden])
Date: 2002-01-25 18:57:09

--- In boost_at_y..., "bill_kempf" <williamkempf_at_h...> wrote:
[...]
> > *IF* the posted design would fully support (and be 
> > compatible wrt C-level cancel/exit/etc functions in 
> > my OLD code) the full functionality (*100%*) of the 
> > following *mandatory* (not optional) POSIX threading 
> > calls (thread/mutex/cond attributes aside for now, 
> > these are just pthread_* calls used in the current 
> > version of mythread.c):
> 
> I don't think it's going to be possible to fully support the C-
level 
> cancel/exit/etc. methods of any "native" API in Boost.Threads, 
though 
> that would be the plan for any proposal to the standard.

Are you saying that pthread_* calls implemented 
on top of some other proprietary platform specific 
primitives/interfaces are NOT "native"? For example, 
is SUN's *Solaris proprietary LWP interfaces* more 
native than SUN's Solaris pthread implementation? 
What is the principle difference with respect to 
MS-WIN proprietary API and some win32 pthread 
implemenation?

> > [***] BTW, the last time I've looked at it,
> > WIN32 version was lacking proper sync. in 
> > boost::condition's destructor.
> 
> Could you send a diff showing the fix you think is required?

Is it helpful enough if I just refer you to the 
comments/description and implementation sections 
of pthread_cond_destroy standard function here:

http://sources.redhat.com/cgi-bin/cvsweb.cgi/pthreads/condvar.c?
rev=1.48&content-type=text/x-cvsweb-markup&cvsroot=pthreads-win32

<?>

regards,
alexander.