[SPv2] Handle transform space reversion in PaintArtifactCompositor |
|||
Issue descriptionThis is probably not easy to fix (but can be worked around) with the current cc interface. Layer transformation and clip should be decoupled from layer hierarchy and we need an interface to associate layers with property nodes directly.
,
Apr 22 2016
The virtual/spv2/fast/clip/010.html started to timeout with https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Mac10.9/builds/30883 so I'm going to update expectations and point to this bug. There are two spv2 CLs in that blamelist - can you please revert if you know what is causing this?
,
Apr 22 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/42a956de9d91711c163ed6b1837689e99829b279 commit 42a956de9d91711c163ed6b1837689e99829b279 Author: kjellander <kjellander@chromium.org> Date: Fri Apr 22 11:43:18 2016 Update WebKit TestExpectations for spv2 and svg layouttests. virtual/spv2/fast/clip/010.html failing on WebKit Mac10.9 (597156) svg/transforms/animated-path-inside-transformed-html.xhtml failing on WebKit Mac10.11 (dbg) (605024) BUG= 597156 , 605024 TBR=trchen@chromium.org, eae@chromium.org Review URL: https://codereview.chromium.org/1911333002 Cr-Commit-Position: refs/heads/master@{#389079} [modify] https://crrev.com/42a956de9d91711c163ed6b1837689e99829b279/third_party/WebKit/LayoutTests/TestExpectations
,
Apr 22 2016
Thanks! The test is expected to crash but somehow Mac builds sit idle after dumping the stack trace. It should be fine to skip the test altogether.
,
Jul 25 2016
,
Sep 13
Layer list mode FTW! |
|||
►
Sign in to add a comment |
|||
Comment 1 by trchen@chromium.org
, Mar 23 2016Example: <div style="position:absolute; clip:rect(0,0,0,0);"> <div style="position:fixed;"></div> </div> The inner element will be clipped by the CSS clip that is based on absolute coordinate, while the element itself paints in frame coordinate. I think this is the only corner case we have.