[Webkit-unassigned] [Bug 96828] webgl/1.0.4/conformance/extensions/oes-vertex-array-object.html fails

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=96828

--- Comment #9 from Kimmo Kinnunen <kkinnunen at apple.com> ---
Comment on attachment 433573
  --> https://bugs.webkit.org/attachment.cgi?id=433573
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=433573&action=review

Thanks for the patch. I'm not sure this approach works for the default case where ANGLE is being used.

Note: supporting WebGL *correctly* on top of OpenGL is very demanding. This is what ANGLE is currently implementing, e.g. most probably in the end the implementation would amount to almost re-implementing ANGLE.
This would indicate that it would maybe be more fruitful to work on getting WPE,GLIB, et al be able to run WebGL with ANGLE. There should not be any architectural problems, but of course it's a bit of a grind.

> Source/WebCore/html/canvas/WebGLVertexArrayObjectOES.cpp:33
> +#include "GLContext.h"

The GLContext.h is an implementation detail of the ports that implement compositor with "texture mapper", e.g. GLIB/WPE/.. , non-apple ports. 

In the WebCore WebGL implementation level, this cannot be used.
We should use GraphicsContextGL abstraction in this level.

I'm not very knowledgeable in the way current non-ANGLE implementation is simulating this VAO feature. However, it probably should follow the structure similar to  what is being done in functions such as WebGLRenderingContextBase::validateVertexAttributes.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20210803/788717a4/attachment-0001.htm>