Bug 68895

Summary:	[Chromium] Layout Test fast/canvas/canvas-composite-transformclip.html is failing
Product:	WebKit	Reporter:	Shinichiro Hamaji <hamaji>
Component:	Tools / Tests	Assignee:	Mike Reed <reed>
Status:	RESOLVED FIXED
Severity:	Normal	CC:	caryclark, epoger, mihaip
Priority:	P2
Version:	528+ (Nightly build)
Hardware:	Unspecified
OS:	Unspecified

Description Shinichiro Hamaji 2011-09-27 06:07:29 PDT

The following layout test is failing on win mac GPU

fast/canvas/canvas-composite-transformclip.html
fast/canvas/canvas-composite.html

Probable cause:

This skia roll? http://src.chromium.org/viewvc/chrome?view=rev&revision=102914

Comment 1 Shinichiro Hamaji 2011-09-27 06:11:28 PDT

Committed r96103: <http://trac.webkit.org/changeset/96103>

Comment 2 Mihai Parparita 2011-09-27 10:30:22 PDT

This was caused by http://crrev.com/102914, which is a Skia roll.

Comment 3 Mike Reed 2011-09-27 10:32:38 PDT

Will be rebaselined tomorrow by today's sheriff.

Comment 4 Shinichiro Hamaji 2011-09-27 22:42:43 PDT

Committed r96191: <http://trac.webkit.org/changeset/96191>

Comment 5 Shinichiro Hamaji 2011-09-27 22:45:38 PDT

Done.

I rebaselined svg/W3C-SVG-1.1/pservers-grad-14-b.svg as well, because the comment says we can rebaseline it after skia 2295.

It seemed Mike the result of fast/canvas/canvas-transforms-during-path.html will change after skia 2330, but it didn't fail actually.

Comment 6 Shinichiro Hamaji 2011-09-28 00:59:51 PDT

Committed r96201: <http://trac.webkit.org/changeset/96201>

Comment 7 Mike Reed 2011-09-28 05:00:27 PDT

Thanks Shinichiro. I didn't trust my local runs of DRT, so I must have included some tests that did in fact not need to be rebaselined. Sorry about that.

Comment 8 Mihai Parparita 2011-09-28 12:18:38 PDT

Looks the there was no rebaselining for the Mac GPU bots. I think this is because webkit-patch rebaseline-expectations didn't know about those bots, I have a patch up at bug 69016 to fix that (and I landed baselines with http://trac.webkit.org/changeset/96244).