Bug 127289 (CVE-2013-6663) - Update overlay scrollbars in single pass
Summary: Update overlay scrollbars in single pass
Alias: CVE-2013-6663
Product: WebKit
Classification: Unclassified
Component: Layout and Rendering (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
Depends on: 127354
  Show dependency treegraph
Reported: 2014-01-20 06:07 PST by Antti Koivisto
Modified: 2014-05-02 14:32 PDT (History)
5 users (show)

See Also:

patch (4.42 KB, patch)
2014-01-20 06:36 PST, Antti Koivisto
andersca: review+
Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Antti Koivisto 2014-01-20 06:07:41 PST
Multi-pass scrollbar resolution is only needed for traditional scrollbars. Overlay scrollbars don't affect layout.
Comment 1 Antti Koivisto 2014-01-20 06:36:06 PST
Created attachment 221656 [details]
Comment 2 WebKit Commit Bot 2014-01-20 06:38:03 PST
Attachment 221656 [details] did not pass style-queue:

ERROR: Source/WebCore/platform/ScrollView.cpp:560:  Should have only a single space after a punctuation in a comment.  [whitespace/comments] [5]
Total errors found: 1 in 2 files

If any of these errors are false positives, please file a bug against check-webkit-style.
Comment 3 Antti Koivisto 2014-01-20 09:17:38 PST
Comment 4 Simon Fraser (smfr) 2014-01-20 12:13:44 PST
Comment on attachment 221656 [details]

I think this might be a no-op after r155660?
Comment 5 WebKit Commit Bot 2014-01-21 09:58:52 PST
Re-opened since this is blocked by bug 127354
Comment 6 Alexey Proskuryakov 2014-01-21 10:37:44 PST
Rolled out in <http://trac.webkit.org/changeset/162447>, and confirmed that this fixed the test.

Looks like this is a good patch overall, but straightening its effects on accessibility seems tricky. It manifested in really weird behavior, see bug 127323.
Comment 7 Alexey Proskuryakov 2014-01-21 11:30:32 PST
> I think this might be a no-op after r155660?

Definitely still not a no-op, because rolling it out changed test behavior.
Comment 8 Simon Fraser (smfr) 2014-01-21 12:24:18 PST
Yeah, I confirmed yesterday that it does eliminate calls to visibleContentsResized()
Comment 9 Antti Koivisto 2014-01-22 05:05:28 PST
Relanded in https://trac.webkit.org/r162515

It tries to resolve accessibility test flakiness by forcing layout in the test. The real bug might be in the AccessibilityController.