Invalidate paint when allocating a DeprecatedPaintLayer.

Invalidate paint when allocating a DeprecatedPaintLayer.

Receiving a DeprecatedPaintLayer can cause the paint invalidation container to
change, so to be correct we must eagerly invalidate before creating the layer.

This will not be needed in Slimming Paint phase 2, because paint invalidation
containers will disappear as a concept.

BUG=461179
Committed: https://crrev.com/91f3f634bef0dc0ed3e3d029001e50497c306a77
git-svn-id: svn://svn.chromium.org/blink/trunk@202304 bbb929c8-8fbe-4397-9dbb-9b2b20218538

[chrishtr, 2015-09-14 23:07:52]

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

[chrishtr, 2015-09-14 23:07:53]

https://codereview.chromium.org/1337233002/diff/20001/LayoutTests/paint/inval...
File LayoutTests/paint/invalidation/invalidate-when-receiving-paint-layer.html
(right):

https://codereview.chromium.org/1337233002/diff/20001/LayoutTests/paint/inval...
LayoutTests/paint/invalidation/invalidate-when-receiving-paint-layer.html:4:
<div id="target" style="opacity: 1; height: 100px; width: 200px;">
First we paint two composited layers: one for the will-change item and one for
the parent of "target" (due to assumed-overlap with the other one). Changing the
opacity of "target" causes it to get a DPL, which in turn causes it to be
squashed into its parent's layer.

However, the paint invalidation container before the change was the parent,
whereas
after receiving the DPL it becomes the HTML element.

[pdr., 2015-09-14 23:29:26]

pdr@chromium.org changed reviewers:
+ pdr@chromium.org

[pdr., 2015-09-14 23:29:27]

https://codereview.chromium.org/1337233002/diff/20001/LayoutTests/paint/inval...
File
LayoutTests/paint/invalidation/invalidate-when-receiving-paint-layer-expected.html
(right):

https://codereview.chromium.org/1337233002/diff/20001/LayoutTests/paint/inval...
LayoutTests/paint/invalidation/invalidate-when-receiving-paint-layer-expected.html:2:
<div style="will-change: transform; position: fixed"></div>
Is this needed?

https://codereview.chromium.org/1337233002/diff/20001/Source/core/layout/Layo...
File Source/core/layout/LayoutBoxModelObject.cpp (right):

https://codereview.chromium.org/1337233002/diff/20001/Source/core/layout/Layo...
Source/core/layout/LayoutBoxModelObject.cpp:292: if
(!RuntimeEnabledFeatures::slimmingPaintV2Enabled() && isRooted())
Can you put the disablers in this block?

[chrishtr, 2015-09-15 19:59:50]

On 2015/09/14 at 23:29:27, pdr wrote:
>
https://codereview.chromium.org/1337233002/diff/20001/LayoutTests/paint/inval...
> File
LayoutTests/paint/invalidation/invalidate-when-receiving-paint-layer-expected.html
(right):
> 
>
https://codereview.chromium.org/1337233002/diff/20001/LayoutTests/paint/inval...
>
LayoutTests/paint/invalidation/invalidate-when-receiving-paint-layer-expected.html:2:
<div style="will-change: transform; position: fixed"></div>
> Is this needed?

Yes, otherwise there are small color differences.

> 
>
https://codereview.chromium.org/1337233002/diff/20001/Source/core/layout/Layo...
> File Source/core/layout/LayoutBoxModelObject.cpp (right):
> 
>
https://codereview.chromium.org/1337233002/diff/20001/Source/core/layout/Layo...
> Source/core/layout/LayoutBoxModelObject.cpp:292: if
(!RuntimeEnabledFeatures::slimmingPaintV2Enabled() && isRooted())
> Can you put the disablers in this block?

Done.

[pdr., 2015-09-15 20:20:32]

LGTM

[chrishtr, 2015-09-15 20:29:18]

The CQ bit was checked by chrishtr@chromium.org

[commit-bot: I haz the power, 2015-09-15 20:29:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1337233002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1337233002/60001

[commit-bot: I haz the power, 2015-09-15 21:01:50]

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

[commit-bot: I haz the power, 2015-09-15 21:01:51]

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_...)

[chrishtr, 2015-09-15 21:45:45]

The CQ bit was checked by chrishtr@chromium.org

[chrishtr, 2015-09-15 21:45:45]

The patchset sent to the CQ was uploaded after l-g-t-m from pdr@chromium.org
Link to the patchset: https://codereview.chromium.org/1337233002/#ps80001
(title: " ")

[commit-bot: I haz the power, 2015-09-15 21:45:55]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1337233002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1337233002/80001

[commit-bot: I haz the power, 2015-09-16 01:30:35]

Committed patchset #5 (id:80001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=202304

[commit-bot: I haz the power, 2015-09-23 12:52:20]

Patchset 5 (id:??) landed as
https://crrev.com/91f3f634bef0dc0ed3e3d029001e50497c306a77