[webkit-reviews] review granted: [Bug 229941] webgl/2.0.y/conformance/extensions/webgl-compressed-texture-s3tc-srgb.html fails on Intel+AMD Metal : [Attachment 437935] Patch

[bugzilla-daemon at webkit.org]

Kenneth Russell <kbr at google.com> has granted Kyle Piddington
<kpiddington at apple.com>'s request for review:
Bug 229941:
webgl/2.0.y/conformance/extensions/webgl-compressed-texture-s3tc-srgb.html
fails on Intel+AMD Metal
https://bugs.webkit.org/show_bug.cgi?id=229941

Attachment 437935: Patch

https://bugs.webkit.org/attachment.cgi?id=437935&action=review




--- Comment #14 from Kenneth Russell <kbr at google.com> ---
Comment on attachment 437935
  --> https://bugs.webkit.org/attachment.cgi?id=437935
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=437935&action=review

Thanks for producing all of the revisions. Looks good to me. r+

> Source/ThirdParty/ANGLE/src/libANGLE/renderer/metal/mtl_utils.mm:136
> +    assert(textureObjFormat.actualAngleFormat().isBlock);

Would prefer ANGLE's ASSERT.

> Source/ThirdParty/ANGLE/src/libANGLE/renderer/metal/mtl_utils.mm:162
> +	       texture->replaceRegion(contextMtl, mtlTextureRegion,
index.getNativeLevel(), layer, buffer.data(), bytesPerRow, 0);

Re-reviewing this, it looks correct for cube map textures because the passed-in
layer is the cube map face index from [0..5].

> LayoutTests/ChangeLog:9
> +	   * TestExpectations:

Comment "Explicitly turn on conformance test until all of webgl/2.0.y is
enabled".