[boost] Re: problem with shared_ptr

From: Alexander Terekhov (terekhov_at_[hidden])
Date: 2003-11-18 09:27:59

Peter Dimov wrote:
> 
> Alexander Terekhov wrote:
> > Peter Dimov wrote:
> > [...]
> >> ... and please report the results. It would also help if you can try
> >> the same test with #define BOOST_LWM_USE_CRITICAL_SECTION.
> >
> > MS CRITICAL_SECTION impl seem to "handoff" lock ownership. 

http://groups.google.com/groups?selm=3f9997eb%241%40news.microsoft.com
("CriticalSections? We have the ownership passing issues of course.")

> >                                                            That's not
> > good.
> 
> Handoff lock ownership? This is code speak for what?

http://groups.google.com/groups?selm=3F5E4104.8D168F8D%40web.de
http://groups.google.com/groups?selm=3f606028%40usenet01.boi.hp.com
(DRB's latest essay on this; read this first)

> 
> > http://groups.google.com/groups?selm=3FA61F96.223FDE85%40web.de
> 
> This looks like a non-broken lightweight_mutex for Win32. 

It was actually meant to replace the pthreads-win32 mutex impl. The
idea was to introduced it in the hypothetical pthreads-win32-and-64 
"derivative".

http://sources.redhat.com/ml/pthreads-win32/2003/msg00108.html
(Subject: mutexes: "food for thought")

However, given rather slow progress (pretty much complete deadlock)
on the fronts of

http://sources.redhat.com/ml/pthreads-win32/2003/msg00111.html
(Subject: License wars)

I mean:

http://sources.redhat.com/ml/pthreads-win32/2003/msg00110.html
(Subject: changing pthreads-win32 license)

I don't expect much progress (on my part) in this respect either.

regards,
alexander.