Don't paint self-painting frameset frames in the layout tree paint recursion.

Don't paint self-painting frameset frames in the layout tree paint recursion.

They are painted via the DeprecatedPaintLayer paint recursion instead.
This previously resulted in double-painting of frameset frames.

BUG=498938
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=197062

[chrishtr, 2015-06-12 18:03:32]

chrishtr@chromium.org changed reviewers:
+ pdr@chromium.org, wkorman@chromium.org

[wkorman, 2015-06-12 19:14:48]

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
File LayoutTests/paint/frames/frameset-with-stacking-contexts.html (right):

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
LayoutTests/paint/frames/frameset-with-stacking-contexts.html:2: <frame
src="resources/subframe.html" style="opacity: 0.6">
Comment or bit of text on what we should see for these to be passing.

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
File LayoutTests/paint/frames/resources/subframe.html (right):

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
LayoutTests/paint/frames/resources/subframe.html:2: Subframe content
Expectations below look like they're from an earlier version of the test, but
maybe that's fine because the rebaselinebot will overwrite them?

https://codereview.chromium.org/1180143003/diff/20001/Source/core/paint/Frame...
File Source/core/paint/FrameSetPainter.cpp (right):

https://codereview.chromium.org/1180143003/diff/20001/Source/core/paint/Frame...
Source/core/paint/FrameSetPainter.cpp:113: LayoutObject* child =
m_layoutFrameSet.firstChild();
It looks like this could be null, at least theoretically, per:

https://code.google.com/p/chromium/codesearch#chromium/src/third_party/WebKit...

If so, add null check?

https://codereview.chromium.org/1180143003/diff/20001/Source/core/paint/Frame...
Source/core/paint/FrameSetPainter.cpp:118: if (child->isLayoutBoxModelObject()
&& toLayoutBoxModelObject(child)->hasSelfPaintingLayer())
Worth adding a comment on why we do this?

[chrishtr, 2015-06-12 19:57:29]

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
File LayoutTests/paint/frames/frameset-with-stacking-contexts.html (right):

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
LayoutTests/paint/frames/frameset-with-stacking-contexts.html:2: <frame
src="resources/subframe.html" style="opacity: 0.6">
On 2015/06/12 at 19:14:48, wkorman wrote:
> Comment or bit of text on what we should see for these to be passing.

Done.

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
File LayoutTests/paint/frames/resources/subframe.html (right):

https://codereview.chromium.org/1180143003/diff/20001/LayoutTests/paint/frame...
LayoutTests/paint/frames/resources/subframe.html:2: Subframe content
On 2015/06/12 at 19:14:48, wkorman wrote:
> Expectations below look like they're from an earlier version of the test, but
maybe that's fine because the rebaselinebot will overwrite them?

Right.

https://codereview.chromium.org/1180143003/diff/20001/Source/core/paint/Frame...
File Source/core/paint/FrameSetPainter.cpp (right):

https://codereview.chromium.org/1180143003/diff/20001/Source/core/paint/Frame...
Source/core/paint/FrameSetPainter.cpp:118: if (child->isLayoutBoxModelObject()
&& toLayoutBoxModelObject(child)->hasSelfPaintingLayer())
On 2015/06/12 at 19:14:48, wkorman wrote:
> Worth adding a comment on why we do this?

Done.

[wkorman, 2015-06-12 20:03:54]

The CQ bit was checked by wkorman@chromium.org

[wkorman, 2015-06-12 20:03:55]

lgtm

[commit-bot: I haz the power, 2015-06-12 20:04:11]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1180143003/40001

[commit-bot: I haz the power, 2015-06-12 21:17:18]

Committed patchset #3 (id:40001) as
https://src.chromium.org/viewvc/blink?view=rev&revision=197062