Implement Paint Containment

Implement Paint Containment

Paint containment acts as a containing block for all objects, including
fixed. It also clips them. It's also a formatting context and a
stacking context.

The approach in this patch differs from the spec in one very notable
way: it does not mess with the computed style of the overflow property.
This is a conscious decision I hope to work with Tab to address.

Paint Containment description from spec:
https://drafts.csswg.org/css-containment/#containment-paint

Intent to implement:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/9W80Kw-z3ss
BUG=561713
Committed: https://crrev.com/472398196d9109be16d13f3af4e8c423ce45d2e3
Cr-Commit-Position: refs/heads/master@{#365178}

[leviw_travelin_and_unemployed, 2015-12-09 00:51:20]

Description was changed from

==========
Implement Paint Containment

Paint Containment description from spec:
https://drafts.csswg.org/css-containment/#containment-paint

Intent to implement:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/9W80Kw-z3ss
BUG=561713
==========

to

==========
Implement Paint Containment

Paint containment acts as a containing block for all objects, including
fixed. It also clips them. It's also a formatting context and a
stacking context.

The approach in this patch differs from the spec in one very notable
way: it does not mess with the computed style of the overflow property.
This is a conscious decision I hope to work with Tab to address.

Paint Containment description from spec:
https://drafts.csswg.org/css-containment/#containment-paint

Intent to implement:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/9W80Kw-z3ss
BUG=561713
==========

[leviw_travelin_and_unemployed, 2015-12-09 00:52:50]

leviw@chromium.org changed reviewers:
+ chrishtr@chromium.org, ojan@chromium.org, pdr@chromium.org,
tabatkins@chromium.org

[leviw_travelin_and_unemployed, 2015-12-09 00:52:51]

PTAL.

Here's a picture of a seal to help with your review: http://imgur.com/tuulpgK

[chrishtr, 2015-12-09 01:41:46]

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/css/resolver/StyleAdjuster.cpp (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/css/resolver/StyleAdjuster.cpp:203: ||
style.containsPaint()))
Does this work with inline elements?

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutBox.cpp (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutBox.cpp:107: ||
!style()->hasAutoZIndex() || style()->shouldCompositeForCurrentAnimations())
!hasAutoZIndex should be enough, line 110 will be a no-op.

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutInline.h (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutInline.h:231: PaintLayerType
layerTypeRequired() const override
I think this method should be un-inlined at this point.

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutInline.h:233: // TODO(leviw): Should
this create an OverflowPaintLayer with containsPaint()?
I think no. Stacking contexts should be self-painting.

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/paint/PaintLayer.h (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/paint/PaintLayer.h:466: return isRootLayer() ||
layerlayoutObject->isPositioned() || hasTransformRelatedProperty() ||
layerlayoutObject->style()->containsPaint();
It's not really positioned..

[leviw_travelin_and_unemployed, 2015-12-09 21:09:24]

PTAL, homeslice.

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/css/resolver/StyleAdjuster.cpp (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/css/resolver/StyleAdjuster.cpp:203: ||
style.containsPaint()))
On 2015/12/09 at 01:41:46, chrishtr wrote:
> Does this work with inline elements?

Yurp yurp yurp. But there were no tests, which is now addressed, so you can
finally rest.

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutBox.cpp (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutBox.cpp:107: ||
!style()->hasAutoZIndex() || style()->shouldCompositeForCurrentAnimations())
On 2015/12/09 at 01:41:46, chrishtr wrote:
> !hasAutoZIndex should be enough, line 110 will be a no-op.

Good point :)

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutInline.h (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutInline.h:231: PaintLayerType
layerTypeRequired() const override
On 2015/12/09 at 01:41:46, chrishtr wrote:
> I think this method should be un-inlined at this point.

Done and...

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutInline.h:233: // TODO(leviw): Should
this create an OverflowPaintLayer with containsPaint()?
On 2015/12/09 at 01:41:46, chrishtr wrote:
> I think no. Stacking contexts should be self-painting.

... done.

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/paint/PaintLayer.h (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/paint/PaintLayer.h:466: return isRootLayer() ||
layerlayoutObject->isPositioned() || hasTransformRelatedProperty() ||
layerlayoutObject->style()->containsPaint();
On 2015/12/09 at 01:41:46, chrishtr wrote:
> It's not really positioned..

That's because this method and its one caller, despite the name, aren't really
looking for a positioned container, they're looking for the container for a
positioned items. I'm not a fan of this code being here to begin with, honestly.

[chrishtr, 2015-12-09 23:34:58]

Could you make some unittest-style layout tests or c++ tests? Example for
inspiration:

https://codereview.chromium.org/1513663002/diff/20001/third_party/WebKit/Layo...

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutObject.h (right):

https://codereview.chromium.org/1490063002/diff/120001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutObject.h:704: bool
hasOverflowClipPaintContainmentOrClip() const { return hasClip() ||
hasOverflowClip() || style()->containsPaint(); }
hasClipRelatedProperty()

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/paint/PaintLayerClipper.cpp (right):

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/paint/PaintLayerClipper.cpp:72:
ASSERT(layoutObject.hasOverflowClip() || layoutObject.hasClip() ||
layoutObject.style()->containsPaint());
hasClipRelatedProperty()

[ojan, 2015-12-10 02:21:41]

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutInline.cpp (right):

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutInline.cpp:1143: || 
style()->hasCompositorProxy() || style()->containsPaint() ? NormalPaintLayer :
NoPaintLayer;
Is there a future world in which we don't need a PaintLayer for contained
elements? It's actually an encouraged use of this feature to put it on all the
block elements on a page, but we wouldn't want to create a PaintLayer for each
of them as that's too expensive.

I was going to ask for a TODO to make it not require a layer, but I wasn't sure
what work that would involve.

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutObject.h (right):

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutObject.h:704: bool
hasOverflowClipPaintContainmentOrClip() const { return hasClip() ||
hasOverflowClip() || style()->containsPaint(); }
Might be time to find a better name for this.

[leviw_travelin_and_unemployed, 2015-12-10 02:45:06]

On 2015/12/10 at 02:21:41, ojan wrote:
>
https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
> File third_party/WebKit/Source/core/layout/LayoutInline.cpp (right):
> 
>
https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
> third_party/WebKit/Source/core/layout/LayoutInline.cpp:1143: || 
style()->hasCompositorProxy() || style()->containsPaint() ? NormalPaintLayer :
NoPaintLayer;
> Is there a future world in which we don't need a PaintLayer for contained
elements? It's actually an encouraged use of this feature to put it on all the
block elements on a page, but we wouldn't want to create a PaintLayer for each
of them as that's too expensive.
> 
> I was going to ask for a TODO to make it not require a layer, but I wasn't
sure what work that would involve.

I worry about how much crazy it would inject into our stacking context code, as
descendants of a contain: paint element would have to consider it even if it
didn't have a layer.

> 
>
https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
> File third_party/WebKit/Source/core/layout/LayoutObject.h (right):
> 
>
https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
> third_party/WebKit/Source/core/layout/LayoutObject.h:704: bool
hasOverflowClipPaintContainmentOrClip() const { return hasClip() ||
hasOverflowClip() || style()->containsPaint(); }
> Might be time to find a better name for this.

Done in my WIP patch.

[chrishtr, 2015-12-10 16:10:52]

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/LayoutInline.cpp (right):

https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/LayoutInline.cpp:1143: || 
style()->hasCompositorProxy() || style()->containsPaint() ? NormalPaintLayer :
NoPaintLayer;
On 2015/12/10 at 02:21:40, ojan wrote:
> Is there a future world in which we don't need a PaintLayer for contained
elements?

PaintLayer may go away in the future in favor of something else, but the things
that make it more heavyweight may or may not.

> It's actually an encouraged use of this feature to put it on all the block
elements on a page, but we wouldn't want to create a PaintLayer for each of them
as that's too expensive.

Well we shouldn't encourage it then...

> 
> I was going to ask for a TODO to make it not require a layer, but I wasn't
sure what work that would involve.

This is a huge undertaking, right? Stacking contexts are implemented by
PaintLayer.

[leviw_travelin_and_unemployed, 2015-12-10 21:56:19]

On 2015/12/10 at 16:10:52, chrishtr wrote:
>
https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
> File third_party/WebKit/Source/core/layout/LayoutInline.cpp (right):
> 
>
https://codereview.chromium.org/1490063002/diff/140001/third_party/WebKit/Sou...
> third_party/WebKit/Source/core/layout/LayoutInline.cpp:1143: || 
style()->hasCompositorProxy() || style()->containsPaint() ? NormalPaintLayer :
NoPaintLayer;
> On 2015/12/10 at 02:21:40, ojan wrote:
> > Is there a future world in which we don't need a PaintLayer for contained
elements?
> 
> PaintLayer may go away in the future in favor of something else, but the
things that make it more heavyweight may or may not.
> 
> > It's actually an encouraged use of this feature to put it on all the block
elements on a page, but we wouldn't want to create a PaintLayer for each of them
as that's too expensive.
> 
> Well we shouldn't encourage it then...
> 
> > 
> > I was going to ask for a TODO to make it not require a layer, but I wasn't
sure what work that would involve.
> 
> This is a huge undertaking, right? Stacking contexts are implemented by
PaintLayer.

Yeah. This is the rub. :(

[leviw_travelin_and_unemployed, 2015-12-10 23:17:39]

PTAL

https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html
(right):

https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html:15:
function test() {
Chrishtr: Is this kinda what you were talking about?

[chrishtr, 2015-12-10 23:31:16]

More tests to add:

1. z-index doesn't escape
2. formatting context
3. clipping and scrolling with otherwise-escaping containing block that no
longer escapes
4. inlines (I think you said you added such a test?)

https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
File
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html
(right):

https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html:15:
function test() {
On 2015/12/10 at 23:17:39, leviw wrote:
> Chrishtr: Is this kinda what you were talking about?

Yeah seem helpful. Please put them in LayoutTests/hittesting/

[chrishtr, 2015-12-11 19:12:57]

On 2015/12/10 at 23:31:16, chrishtr wrote:
> More tests to add:
> 
> 1. z-index doesn't escape
> 2. formatting context
> 3. clipping and scrolling with otherwise-escaping containing block that no
longer escapes
> 4. inlines (I think you said you added such a test?)

Alternatively, you could add a unittest inheriting from RenderingTest that
verifies when adding contains:paint that the resulting object is a formatting
context, containing block for everything,
and a stacking context. I think that would actually be a better way to do it
than adding more layout tests, since it's more targeted.

> 
>
https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
> File
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html
(right):
> 
>
https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
>
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html:15:
function test() {
> On 2015/12/10 at 23:17:39, leviw wrote:
> > Chrishtr: Is this kinda what you were talking about?
> 
> Yeah seem helpful. Please put them in LayoutTests/hittesting/

[leviw_travelin_and_unemployed, 2015-12-11 23:27:36]

PTAL

On 2015/12/11 at 19:12:57, chrishtr wrote:
> On 2015/12/10 at 23:31:16, chrishtr wrote:
> > More tests to add:
> > 
> > 1. z-index doesn't escape
> > 2. formatting context
> > 3. clipping and scrolling with otherwise-escaping containing block that no
longer escapes
> > 4. inlines (I think you said you added such a test?)
I converted all the existing tests to use have both an inline and block version.

> 
> Alternatively, you could add a unittest inheriting from RenderingTest that
verifies when adding contains:paint that the resulting object is a formatting
context, containing block for everything,
> and a stacking context. I think that would actually be a better way to do it
than adding more layout tests, since it's more targeted.

Done.

> 
> > 
> >
https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
> > File
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html
(right):
> > 
> >
https://codereview.chromium.org/1490063002/diff/180001/third_party/WebKit/Lay...
> >
third_party/WebKit/LayoutTests/fast/css/containment/paint-containment-hittest.html:15:
function test() {
> > On 2015/12/10 at 23:17:39, leviw wrote:
> > > Chrishtr: Is this kinda what you were talking about?
> > 
> > Yeah seem helpful. Please put them in LayoutTests/hittesting/

[chrishtr, 2015-12-11 23:39:19]

lgtm

https://codereview.chromium.org/1490063002/diff/240001/third_party/WebKit/Sou...
File third_party/WebKit/Source/core/layout/PaintContainmentTest.cpp (right):

https://codereview.chromium.org/1490063002/diff/240001/third_party/WebKit/Sou...
third_party/WebKit/Source/core/layout/PaintContainmentTest.cpp:21:
EXPECT_TRUE(obj.hasClipRelatedProperty());
EXPECT_TRUE(obj.style()->containsPaint())

[leviw_travelin_and_unemployed, 2015-12-12 00:38:11]

The CQ bit was checked by leviw@chromium.org

[leviw_travelin_and_unemployed, 2015-12-12 00:38:11]

The patchset sent to the CQ was uploaded after l-g-t-m from
chrishtr@chromium.org
Link to the patchset: https://codereview.chromium.org/1490063002/#ps260001
(title: "Add containsPaint check")

[commit-bot: I haz the power, 2015-12-12 00:39:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1490063002/260001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1490063002/260001

[commit-bot: I haz the power, 2015-12-12 00:58:33]

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

[commit-bot: I haz the power, 2015-12-12 00:58:34]

Try jobs failed on following builders:
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[leviw_travelin_and_unemployed, 2015-12-12 01:18:52]

The CQ bit was checked by leviw@chromium.org

[leviw_travelin_and_unemployed, 2015-12-12 01:18:53]

The patchset sent to the CQ was uploaded after l-g-t-m from
chrishtr@chromium.org
Link to the patchset: https://codereview.chromium.org/1490063002/#ps280001
(title: "Add missing semicolon")

[commit-bot: I haz the power, 2015-12-12 01:23:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1490063002/280001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1490063002/280001

[commit-bot: I haz the power, 2015-12-12 02:56:42]

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

[commit-bot: I haz the power, 2015-12-12 02:56:43]

Try jobs failed on following builders:
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[leviw_travelin_and_unemployed, 2015-12-14 20:21:40]

The CQ bit was checked by leviw@chromium.org

[leviw_travelin_and_unemployed, 2015-12-14 20:21:40]

The patchset sent to the CQ was uploaded after l-g-t-m from
chrishtr@chromium.org
Link to the patchset: https://codereview.chromium.org/1490063002/#ps300001
(title: "Fix include patch")

[commit-bot: I haz the power, 2015-12-14 20:22:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1490063002/300001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1490063002/300001

[commit-bot: I haz the power, 2015-12-14 20:25:47]

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

[commit-bot: I haz the power, 2015-12-14 20:25:48]

Try jobs failed on following builders:
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
  mac_chromium_compile_dbg_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_comp...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)

[leviw_travelin_and_unemployed, 2015-12-14 20:31:55]

The CQ bit was checked by leviw@chromium.org

[leviw_travelin_and_unemployed, 2015-12-14 20:31:55]

The patchset sent to the CQ was uploaded after l-g-t-m from
chrishtr@chromium.org
Link to the patchset: https://codereview.chromium.org/1490063002/#ps320001
(title: "Rebased")

[commit-bot: I haz the power, 2015-12-14 20:33:28]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1490063002/320001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1490063002/320001

[commit-bot: I haz the power, 2015-12-14 20:57:05]

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

[commit-bot: I haz the power, 2015-12-14 20:57:07]

Try jobs failed on following builders:
  chromeos_amd64-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_amd64-...)

[leviw_travelin_and_unemployed, 2015-12-14 22:17:50]

The CQ bit was checked by leviw@chromium.org

[leviw_travelin_and_unemployed, 2015-12-14 22:17:50]

The patchset sent to the CQ was uploaded after l-g-t-m from
chrishtr@chromium.org
Link to the patchset: https://codereview.chromium.org/1490063002/#ps340001
(title: "fix merge")

[commit-bot: I haz the power, 2015-12-14 22:21:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1490063002/340001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1490063002/340001

[commit-bot: I haz the power, 2015-12-15 03:51:41]

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

[commit-bot: I haz the power, 2015-12-15 03:51:42]

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

[leviw_travelin_and_unemployed, 2015-12-15 05:29:10]

The CQ bit was checked by leviw@chromium.org

[commit-bot: I haz the power, 2015-12-15 05:32:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1490063002/340001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1490063002/340001

[commit-bot: I haz the power, 2015-12-15 06:20:22]

Committed patchset #18 (id:340001)

[commit-bot: I haz the power, 2015-12-15 06:21:30]

Description was changed from

==========
Implement Paint Containment

Paint containment acts as a containing block for all objects, including
fixed. It also clips them. It's also a formatting context and a
stacking context.

The approach in this patch differs from the spec in one very notable
way: it does not mess with the computed style of the overflow property.
This is a conscious decision I hope to work with Tab to address.

Paint Containment description from spec:
https://drafts.csswg.org/css-containment/#containment-paint

Intent to implement:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/9W80Kw-z3ss
BUG=561713
==========

to

==========
Implement Paint Containment

Paint containment acts as a containing block for all objects, including
fixed. It also clips them. It's also a formatting context and a
stacking context.

The approach in this patch differs from the spec in one very notable
way: it does not mess with the computed style of the overflow property.
This is a conscious decision I hope to work with Tab to address.

Paint Containment description from spec:
https://drafts.csswg.org/css-containment/#containment-paint

Intent to implement:
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/9W80Kw-z3ss
BUG=561713

Committed: https://crrev.com/472398196d9109be16d13f3af4e8c423ce45d2e3
Cr-Commit-Position: refs/heads/master@{#365178}
==========

[commit-bot: I haz the power, 2015-12-15 06:21:31]

Patchset 18 (id:??) landed as
https://crrev.com/472398196d9109be16d13f3af4e8c423ce45d2e3
Cr-Commit-Position: refs/heads/master@{#365178}