[webkit-reviews] review granted: [Bug 131867] [WebGL] Limit maximum texture sizes on older Intel hardware : [Attachment 229686] Patch

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Fri Apr 18 17:17:08 PDT 2014


Geoffrey Garen <ggaren at apple.com> has granted Dean Jackson <dino at apple.com>'s
request for review:
Bug 131867: [WebGL] Limit maximum texture sizes on older Intel hardware
https://bugs.webkit.org/show_bug.cgi?id=131867

Attachment 229686: Patch
https://bugs.webkit.org/attachment.cgi?id=229686&action=review

------- Additional Comments from Geoffrey Garen <ggaren at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=229686&action=review


r=me

> Source/WebCore/platform/graphics/opengl/Extensions3DOpenGLCommon.cpp:103
> +    // Intel HD 3000 devices have problems with
> +    // large textures. <rdar://problem/16649140>

This can be one line.

> Source/WebCore/platform/graphics/opengl/GraphicsContext3DOpenGL.cpp:252
> +    case MAX_TEXTURE_SIZE:
> +	   ::glGetIntegerv(MAX_TEXTURE_SIZE, value);
> +	   if (getExtensions()->requiresRestrictedMaximumTextureSize())
> +	       *value = std::min(4096, *value);
> +	   break;
> +    case MAX_CUBE_MAP_TEXTURE_SIZE:
> +	   ::glGetIntegerv(MAX_CUBE_MAP_TEXTURE_SIZE, value);
> +	   if (getExtensions()->requiresRestrictedMaximumTextureSize())
> +	       *value = std::min(1024, *value);
> +	   break;

Should we really silently truncate instead of just failing and returning an
error? What's the expected behavior here?


More information about the webkit-reviews mailing list