Set geometry dirty bits on composited layer mappings when updating squashed layers.

Set geometry dirty bits on composited layer mappings when updating squashed layers.

The lack of setting geometry dirty bits in these cases is wrong. However, at the moment it is masked by setting needs geometry dirty bits more globally upwards in the system. When that is cleaned up these targeted ones will be in place.

BUG=

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169395

[abarth-chromium, 2014-03-17 17:17:03]

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
File LayoutTests/compositing/squashing/squashed-layer-color-update-expected.txt
(right):

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
LayoutTests/compositing/squashing/squashed-layer-color-update-expected.txt:9:
RenderBlock (positioned) {DIV} at (8,90) size 40x40 [bgcolor=#008000]
Can this be a ref test instead of a render tree dump (e.g., use an
-expected.html file)?

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
File LayoutTests/compositing/squashing/squashed-layer-color-update.html (right):

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
LayoutTests/compositing/squashing/squashed-layer-color-update.html:29:
setTimeout(function() {
Why set timeout?  If you're trying to synchronize with the rendering system, you
might want to use requestAnimationFrame.

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
File Source/core/rendering/compositing/CompositedLayerMapping.cpp (right):

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
Source/core/rendering/compositing/CompositedLayerMapping.cpp:2067:
setNeedsGeometryUpdate();
Should we move this call inside setGroupedMapping?  Don't if the grouped mapping
was non-zero before, don't we need to dirty the old composited layer mapping
too?

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
File Source/core/rendering/compositing/RenderLayerCompositor.cpp (right):

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
Source/core/rendering/compositing/RenderLayerCompositor.cpp:659: bool
RenderLayerCompositor::updateSquashingAssignment(RenderLayer* layer,
SquashingState& squashingState, const CompositingStateTransitionType
compositedLayerUpdate)
What's the difference between these two updateSquashingLayerAssignment
functions?

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
Source/core/rendering/compositing/RenderLayerCompositor.cpp:700:
layer->setGroupedMapping(0);
Maybe this call to setNeedsGeometryUpdate should happen inside
setGroupedMapping.  Presumably every caller of setGroupedMapping needs this as
well.

[chrishtr, 2014-03-17 18:10:54]

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
File LayoutTests/compositing/squashing/squashed-layer-color-update-expected.txt
(right):

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
LayoutTests/compositing/squashing/squashed-layer-color-update-expected.txt:9:
RenderBlock (positioned) {DIV} at (8,90) size 40x40 [bgcolor=#008000]
On 2014/03/17 17:17:03, abarth wrote:
> Can this be a ref test instead of a render tree dump (e.g., use an
> -expected.html file)?

Done.

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
File LayoutTests/compositing/squashing/squashed-layer-color-update.html (right):

https://codereview.chromium.org/200623006/diff/60001/LayoutTests/compositing/...
LayoutTests/compositing/squashing/squashed-layer-color-update.html:29:
setTimeout(function() {
On 2014/03/17 17:17:03, abarth wrote:
> Why set timeout?  If you're trying to synchronize with the rendering system,
you
> might want to use requestAnimationFrame.

Done.

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
File Source/core/rendering/compositing/CompositedLayerMapping.cpp (right):

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
Source/core/rendering/compositing/CompositedLayerMapping.cpp:2067:
setNeedsGeometryUpdate();
On 2014/03/17 17:17:03, abarth wrote:
> Should we move this call inside setGroupedMapping?  Don't if the grouped
mapping
> was non-zero before, don't we need to dirty the old composited layer mapping
> too?

Done.

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
File Source/core/rendering/compositing/RenderLayerCompositor.cpp (right):

https://codereview.chromium.org/200623006/diff/60001/Source/core/rendering/co...
Source/core/rendering/compositing/RenderLayerCompositor.cpp:700:
layer->setGroupedMapping(0);
On 2014/03/17 17:17:03, abarth wrote:
> Maybe this call to setNeedsGeometryUpdate should happen inside
> setGroupedMapping.  Presumably every caller of setGroupedMapping needs this as
> well.

Done.

[abarth-chromium, 2014-03-17 19:04:31]

lgtm

[chrishtr, 2014-03-17 19:39:14]

The CQ bit was checked by chrishtr@chromium.org

[commit-bot: I haz the power, 2014-03-17 19:39:22]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/chrishtr@chromium.org/200623006/110001

[commit-bot: I haz the power, 2014-03-17 20:23:03]

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

[commit-bot: I haz the power, 2014-03-17 20:23:04]

Try jobs failed on following builders:
  tryserver.blink on win_blink_rel

[chrishtr, 2014-03-17 21:31:18]

The CQ bit was checked by chrishtr@chromium.org

[commit-bot: I haz the power, 2014-03-17 21:31:34]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/chrishtr@chromium.org/200623006/110001

[commit-bot: I haz the power, 2014-03-17 22:28:59]

Change committed as 169395