Bug 45415 - [chromium] Always do 2d canvas clearRect() in hardware
Summary: [chromium] Always do 2d canvas clearRect() in hardware
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Other OS X 10.5
: P2 Normal
Assignee: James Robinson
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-09-08 14:51 PDT by James Robinson
Modified: 2010-09-08 18:49 PDT (History)
5 users (show)

See Also:

Attachments
Patch (2.03 KB, patch)
2010-09-08 14:57 PDT, James Robinson 		kbr: review+
Details | Formatted Diff | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description James Robinson 2010-09-08 14:51:19 PDT 
[chromium] Always do 2d canvas clearRect() in hardware
Comment 1 James Robinson 2010-09-08 14:57:50 PDT 
Created attachment 66946 [details]
Patch
Comment 2 James Robinson 2010-09-08 14:58:46 PDT 
This patch fixes fuzz on the edge of the FPS meter of the IE test drive pages caused by the end of the white FPS bar extending past the edge of the gradient.
Comment 3 James Robinson 2010-09-08 16:39:58 PDT 
kling in IRC pointed out that we should _not_ apply a shadow to clearRect(), even though CoreGraphics does.

There's still the issue that clearRect() is supposed to only clear within the active clipping region (this overclears), so this isn't patch isn't perfect, but it's better than what we do currently.
Comment 4 Kenneth Russell 2010-09-08 16:49:26 PDT 
Comment on attachment 66946 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=66946&action=prettypatch

Looks OK.
Comment 5 James Robinson 2010-09-08 16:54:56 PDT 
Committed r67039: <http://trac.webkit.org/changeset/67039>