[chromium] Clamp dirty regions on content layers to layer bounds
Created attachment 90266 [details] Patch
This is the WebKit fix for http://crbug.com/79749.
Comment on attachment 90266 [details] Patch unofficial r+ although it probably needs a test too...
Comment on attachment 90266 [details] Patch Agree re: test. We're likely to be mucking about with this code a good bit over the next few weeks and I don't want to rewrite the same bug again without a test.
Because this manually fixes both the original bug and another test case that Vangelis found, I'm going to commit this but leave the bug open until I write a test case.
(In reply to comment #5) > Because this manually fixes both the original bug and another test case that Vangelis found, I'm going to commit this but leave the bug open until I write a test case. Sounds good to me (unofficially). Thanks very much for the quick fix!
Committed r84529: <http://trac.webkit.org/changeset/84529>
Good news, everybody! We already had three test cases that demonstrated this bug. http://test-results.appspot.com/dashboards/flakiness_dashboard.html#showExpectations=true&tests=media%2Fcontrols-styling.html%2Cmedia%2Fvideo-volume-slider.html%2Cmedia%2Fvideo-zoom-controls.html&group=%40ToT%20GPU%20Mesa%20-%20chromium.org ...they were just all marked as IMAGE fail.