Don't use cached subsequence when clip changes

Don't use cached subsequence when clip changes

Previously we disable subsequence caching for layers having clip or
overflow-clip to ensure descendant layers to be repainted when an
ancestor layer's clip changes. This doesn't work if a descendant layer
is painted not using the same PaintLayerPaintingInfo of the ancestor
layer (e.g. when the ancestor layer is not a stacking context). This
also unnecessarily disables subsequence caching even if clipping
doesn't change.

Now save previous paintingClipRects and check if it changes before
issuing subsequence display item.

BUG=533717, 553392
TEST=paint/invalidation/repaint-subsequence-on-ancestor-clip-change.html
TEST=paint/invalidation/repaint-subsequence-on-ancestor-clip-change-complex.html

Committed: https://crrev.com/1a67416323f77edc8c06f17151bac351fba69e11
Cr-Commit-Position: refs/heads/master@{#360535}

[Xianzhu, 2015-11-18 00:03:34]

wangxianzhu@chromium.org changed reviewers:
+ chrishtr@chromium.org

[chrishtr, 2015-11-18 00:32:02]

https://codereview.chromium.org/1452313004/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/paint/PaintLayerClipper.cpp (right):

https://codereview.chromium.org/1452313004/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/paint/PaintLayerClipper.cpp:154: void
PaintLayerClipper::clearClipRectsIncludingDescendants()
It looks like PaintLayer::updateLayerPositionsAfterLayout() calls this
unconditionally?

[Xianzhu, 2015-11-18 22:37:05]

Description was changed from

==========
Set layer needs repaint when invalidating clip

Previously we disable subsequence caching for layers having clip or
overflow-clip to ensure descendant layers to be repainted when an
ancestor layer's clip changes. This doesn't work if a descendant layer
is painted not using the same PaintLayerPaintingInfo of the ancestor
layer (e.g. when the ancestor layer is not a stacking context). This
also unnecessarily disables subsequence caching even if clipping
doesn't change.

Now call PaintLayer::setNeedsRepaint() when PaintLayerClipper cache
is invalidated (which occurs for all descendant layers when an
ancestor's clipping changes).

BUG=533717,553392
TEST=paint/invalidation/repaint-subsequence-on-ancestor-clip-change.html
TEST=paint/invalidation/repaint-subsequence-on-ancestor-clip-change-complex.html
==========

to

==========
Don't use cached subsequence when clip changes

Previously we disable subsequence caching for layers having clip or
overflow-clip to ensure descendant layers to be repainted when an
ancestor layer's clip changes. This doesn't work if a descendant layer
is painted not using the same PaintLayerPaintingInfo of the ancestor
layer (e.g. when the ancestor layer is not a stacking context). This
also unnecessarily disables subsequence caching even if clipping
doesn't change.

Now save previous paintingClipRects and check if it changes before
issuing subsequence display item.

BUG=533717,553392
TEST=paint/invalidation/repaint-subsequence-on-ancestor-clip-change.html
TEST=paint/invalidation/repaint-subsequence-on-ancestor-clip-change-complex.html
==========

[Xianzhu, 2015-11-18 22:38:15]

https://codereview.chromium.org/1452313004/diff/1/third_party/WebKit/Source/c...
File third_party/WebKit/Source/core/paint/PaintLayerClipper.cpp (right):

https://codereview.chromium.org/1452313004/diff/1/third_party/WebKit/Source/c...
third_party/WebKit/Source/core/paint/PaintLayerClipper.cpp:154: void
PaintLayerClipper::clearClipRectsIncludingDescendants()
On 2015/11/18 00:32:02, chrishtr wrote:
> It looks like PaintLayer::updateLayerPositionsAfterLayout() calls this
> unconditionally?

You are right.

The latest patch checks actual change of painting clip rects.

[chrishtr, 2015-11-18 23:07:17]

The CQ bit was checked by chrishtr@chromium.org

[chrishtr, 2015-11-18 23:07:17]

lgtm

[commit-bot: I haz the power, 2015-11-18 23:09:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1452313004/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1452313004/40001

[commit-bot: I haz the power, 2015-11-19 02:15:16]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-11-19 02:15:17]

Try jobs failed on following builders:
  linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[Xianzhu, 2015-11-19 04:42:50]

The CQ bit was checked by wangxianzhu@chromium.org

[commit-bot: I haz the power, 2015-11-19 04:43:48]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1452313004/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1452313004/40001

[commit-bot: I haz the power, 2015-11-19 06:22:05]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-11-19 06:23:10]

Patchset 3 (id:??) landed as
https://crrev.com/1a67416323f77edc8c06f17151bac351fba69e11
Cr-Commit-Position: refs/heads/master@{#360535}