[chromium] Content and render surface textures not freed when compositor context destroyed
Created attachment 104563 [details] Patch
We talked some in person about how to get better automated testing for stuff like this. I think the general thinking is that we need a way to test the compositing path with either a mock GraphicsContext3D or something lower-level to verify that we're actually releasing all of the resources we create.
Unofficially, looks good to me.
Comment on attachment 104563 [details] Patch rs=me
Comment on attachment 104563 [details] Patch Clearing flags on attachment: 104563 Committed r93523: <http://trac.webkit.org/changeset/93523>
All reviewed patches have been landed. Closing bug.