Summary: | WebGL: Move the format conversion for 16-bit per channel formats into Core Graphics port only | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Jun Jiang <jun.a.jiang> | ||||
Component: | WebGL | Assignee: | Jun Jiang <jun.a.jiang> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | dino, kbr, webkit.review.bot | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
Attachments: |
|
Description
Jun Jiang
2013-01-30 00:39:41 PST
Created attachment 185417 [details]
Patch
Comment on attachment 185417 [details]
Patch
Very nice work and simplification. Before setting the CQ bit, was this tested with all of the existing layout tests, WebGL conformance tests, and Mozilla's WebGLPerf test suite?
Hi, Kenneth. I used to test this patch using the existing Layout tests, Mozilla's WebGLPerf tests and WebGL 1.0.1 conformance tests. And it was OK. Today I tried the latest WebGL conformance test in https://www.khronos.org/registry/webgl/sdk/tests/webgl-conformance-tests.html(newer than version 1.0.1), and it met some issue(RGB8->RGBA32F conversion) for CG port when running safari on MAC. And it was related with another bug and I had submitted fix at https://bugs.webkit.org/show_bug.cgi?id=108812. Comment on attachment 185417 [details]
Patch
OK, sounds fine. Since the other bug is preexisting, doing these fixes separately sounds best. Thanks for the patch.
Comment on attachment 185417 [details] Patch Clearing flags on attachment: 185417 Committed r141841: <http://trac.webkit.org/changeset/141841> All reviewed patches have been landed. Closing bug. |