Hi Robert,

Thanks for the example, with this I was able to reproduce the problem and
confirm fix with your changes, and eventually with my changes that add a
State::releaseGLObjects() and a ShaderComposer::releaseGLObjects(),  As the
"official" fix using releaseGLObjects() changes the ABI I have only merged
this with svn/trunk, with the OSG-3.2 branch I have merged your changes as
this workarounds the problem without changing the ABI.  Changes to
svn/trunk and OSG-3.2 branch now checked in.

Cheers,
Robert.


On 21 October 2013 14:07, Robert Milharcic <[email protected]>wrote:

> On 21.10.2013 12:13, Robert Osfield wrote:
>
>> I have just done another review of the changes and am currently inclined
>> to
>> implement a releaseGLObjects() method in osg::State in line with how the
>> scene graph objects do this.  This would be called prior to the
>> osg::deleteAllGLObjects() call.  With this in place I don't think the
>> addition of the _state = 0 will be required.
>>
>> Thoughts?
>>
>
> Hi Robert,
>
> Agreed. Implementing releaseGLObjects() is a way to go. I'm attaching one
> of the osg examples modified in a way to reproduce the problem on most
> drivers and graphics adapters.
>
> Best Regards
> Robert MIlharcic
>
> _______________________________________________
> osg-submissions mailing list
> [email protected]
>
> http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org
>
>
_______________________________________________
osg-submissions mailing list
[email protected]
http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org

Reply via email to