[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Tue, 24 Jun 2003 14:09:09 +0530
If memory serves me right, Thong (Tum) Nguyen wrote:
Amazing Tum ! ... we postpone it for ~5 months because it's tough and you
finish it in under a week !!!!
> Yah, finally submitted the patch :-).
For the many headed who are still looking for a place to download this from,
this is in http://savannah.gnu.org/projects/dotgnu-pnet/ in the patches
section .. Patch #1653 and Patch #1654 .
> I've spent the last couple of sleepless days trying to debug some
> problems with the threading support that popped up when I installed the
> GC. It seems pretty stable now.
I am tempted to take your word for it ... but I'd rather have a couple
of seasoned .NET guys have a pop at it and see how it can be pushed
and poked :-)
Testers on Win32 , OS X and *BSD have to report success before we can
commit this patch .. So anyone ?
> I've implemented two different types of monitors: object header based
> and hashtable based with object-header-based the default. You can
> enable hashtable based monitors by compiling with USE_HASHING_MONITORS.
> Hashtable based monitors appear to be around 3 times slower which isn't
> really that bad considering it'll save 4 bytes per object :-).
/me thinks he likes Object header based stuff as we already have the Object
in the cache when we're working on it ... (looking at the Parrot design,
they seem to see cache fluff as an important speed criterion) .
> The Thread, WaitHandle and Monitor objects now all work :D
/me starts patching ....
The difference between insanity and genius is measured by success
- [DotGNU]Threading, Thong \(Tum\) Nguyen, 2003/06/23
- Re: [DotGNU]Threading,
Gopal V <=