layers-inside-overflow-scroll.html overflow-compositing-descendant.html scroll-ancestor-update.html self-painting-layers.html See http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=compositing%2Flayers-inside-overflow-scroll.html%2Ccompositing%2Foverflow%2Foverflow-compositing-descendant.html%2Ccompositing%2Foverflow%2Fscroll-ancestor-update.html%2Ccompositing%2Fself-painting-layers.html From the time that the failures started, it looks like this is due to http://trac.webkit.org/changeset/100360, perhaps due to unexpected fallback paths for expected results?
Committed r100430: <http://trac.webkit.org/changeset/100430>
They just needed baselines: http://trac.webkit.org/changeset/100557
Some tests are still failing on 10.6 ... compositing/layer-creation/spanOverlapsCanvas.html compositing/layers-inside-overflow-scroll.html compositing/self-painting-layers.html compositing/video-page-visibility.html compositing/geometry/video-opacity-overlay.html compositing/geometry/video-fixed-scrolling.html compositing/geometry/clipped-video-controller.html compositing/overflow/scroll-ancestor-update.html compositing/overflow/overflow-compositing-descendant.html http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=compositing%2Flayer-creation%2FspanOverlapsCanvas.html%2Ccompositing%2Fgeometry%2Fclipped-video-controller.html%2Ccompositing%2Fgeometry%2Fvideo-fixed-scrolling.html%2Ccompositing%2Fgeometry%2Fvideo-opacity-overlay.html%2Ccompositing%2Flayers-inside-overflow-scroll.html%2Ccompositing%2Foverflow%2Foverflow-compositing-descendant.html%2Ccompositing%2Foverflow%2Fscroll-ancestor-update.html%2Ccompositing%2Fself-painting-layers.html%2Ccompositing%2Fvideo-page-visibility.html
Committed r100598: <http://trac.webkit.org/changeset/100598>
These tests are still failing: compositing/geometry/video-opacity-overlay.html compositing/geometry/video-fixed-scrolling.html compositing/overflow/overflow-compositing-descendant.html compositing/self-painting-layers.html compositing/overflow/scroll-ancestor-update.html compositing/geometry/clipped-video-controller.html Were the baselines wrong?
'598 just added expectations, not baselines. It looks like some of these (at least compositing/geometry/video-opacity-overlay.html) are in fact rendering correctly but need a rebaseline. They've been somewhat unstable lately because of recent configuration changes which changed the tests from using the software video path to using H/W colorspace conversion and scaling, which very slightly changes the video's display.
As far as I can tell, they all seem to just need a rebaseline. Steve/James, can you please verify that and submit new baselines? Thanks!
I filed the bug as WebKit gardener - leaving to James to confirm as I think he's the owner of these tests.
http://trac.webkit.org/changeset/100866