Hi Gerrit,
Gerrit Voss wrote:
> sorry I was a little busy and hadn't have time to follow it to closely.
no worries, you saved the day just when I was running out of ideas :)
> On Thu, 2009-01-29 at 18:34 -0600, Carsten Neumann wrote:
>> There is something strange going on. I've run aspect1render on linux
>> under bugle and it shows that only one thread makes OpenGL (and GLX)
>> calls (and it is the expected thread running on aspect 1 - checked by
>> introducing some nonsense calls from aspect 0).
>> Running under gDEBugger crashes the program and partly gDEBugger (it
>> does not allow me to stop the crashes process).
>> Finally running on windows (32 bit) crashes as well, I've not looked at
>> it in detail yet, will do so over the next days.
>>
> small hint, you create you GL context in one thread and that draw from
> another. Move the initAspect1 call to runAspect1 instead of doing it
> from a separate run call and things seem to work as under linux.
oh, are you saying each runFunction call creates a new OS thread even if
issued from the same OSG::Thread object? Hm, now that I think about it,
makes sense of course.
Ok, so the fix would be to have just one function that is the union of
initAspect1 and runAspect1 and replace the synchronization currently
done with the join() by just waiting at the barrier.
Thanks,
Carsten
------------------------------------------------------------------------------
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
_______________________________________________
Opensg-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/opensg-users
