Hi Js,

I think the testing should be sufficient, and to warrent merging the change
and to look for wider.

I'm not at my computer right (using my phone) now so will merge tomorrow

On 1 Dec 2010 20:18, "Jean-Sébastien Guay" <[email protected]>
wrote:

Hi Robert,



>> The answer though, is it does work, in the pbuffertest at least. Removing
>> the makeCurrentImp...
No, but at least it confirms that changing my submitted file by this would
fix the observed problem that realizeImplementation() did a makeCurrent and
then the GraphicsThread couldn't makeCurrent() successfully after that. So
it's a partial test, not a complete one. :-)



> running osgprerender or osgprerendercubemap with --pbufer should force
> the use of wglShareList...
I ran osgprerendercubemap --pbuffer, with a breakpoint at
PixelBufferWin32.cpp line 664 (the call to wglShareLists()). The breakpoint
was hit multiple times, and never caused any error (the OSG_NOTICE line just
below). And the visual result (the reflection on the central sphere) looked
correct.

Is this sufficient testing to be confident the removal of
makeCurrentImplementation() in realizeImplementation() is safe?



Thanks,

J-S
-- 
______________________________________________________
Jean-Sebastien Guay    jea...
_______________________________________________
osg-submissions mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org

Reply via email to