Always give RenderParts a RenderLayer.

Always give RenderParts a RenderLayer.

This lays the groundwork for removing scheduleLayerUpdate calls in
RenderLayerCompositor for iframes. Instead of changing the layer
type when the frame is composited, change it so that the layer
type remains ForcedLayer, but the isSelfPainting bit respects
the m_compositing bit on RenderLayerCompositor.

This exposed a bug in RenderBox::nodeAtPoint where it wouldn't
recurse correctly because the child now has a layer, but isn't
self-painting. Before we'd never hit this code path with
a non-self-painting layer.

The rebaselines are all just cases of things getting a layer
where they didn't used to.

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=175436

[ojan, 2014-05-29 23:48:58]

The NeedsRebaseline lines conflict with some NeedsManualRebaseline lines, so
don't CQ. I'm hoping they'll be resolved soon and I can just commit the proper
thing. But, this is ready for review.

[ojan, 2014-06-03 03:08:45]

ping. I'd like to land this so I can test
https://codereview.chromium.org/302093002/ on the trybot.

[esprehn, 2014-06-03 03:49:17]

lgtm

https://codereview.chromium.org/302083002/diff/1/Source/core/rendering/Render...
File Source/core/rendering/RenderEmbeddedObject.cpp (right):

https://codereview.chromium.org/302083002/diff/1/Source/core/rendering/Render...
Source/core/rendering/RenderEmbeddedObject.cpp:69: // This just use
RenderPart::layerTypeRequired, because RenderLayerCompositor
typo, -> uses

https://codereview.chromium.org/302083002/diff/1/Source/core/rendering/Render...
Source/core/rendering/RenderEmbeddedObject.cpp:69: // This just use
RenderPart::layerTypeRequired, because RenderLayerCompositor
typo, -> uses

[commit-bot: I haz the power, 2014-06-03 03:50:09]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ojan@chromium.org/302083002/1

[commit-bot: I haz the power, 2014-06-03 03:50:29]

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

[commit-bot: I haz the power, 2014-06-03 03:50:30]

Failed to apply patch for Source/core/rendering/RenderEmbeddedObject.cpp:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file Source/core/rendering/RenderEmbeddedObject.cpp
  Hunk #1 FAILED at 66.
  1 out of 1 hunk FAILED -- saving rejects to file
Source/core/rendering/RenderEmbeddedObject.cpp.rej

Patch:       Source/core/rendering/RenderEmbeddedObject.cpp
Index: Source/core/rendering/RenderEmbeddedObject.cpp
diff --git a/Source/core/rendering/RenderEmbeddedObject.cpp
b/Source/core/rendering/RenderEmbeddedObject.cpp
index
a4aa641643af09ebcd0395a9f07376d804dda4fd..08e3276a8af9aabac22e0d447196ca1ed4f211ea
100644
--- a/Source/core/rendering/RenderEmbeddedObject.cpp
+++ b/Source/core/rendering/RenderEmbeddedObject.cpp
@@ -66,11 +66,12 @@ RenderEmbeddedObject::~RenderEmbeddedObject()
 
 LayerType RenderEmbeddedObject::layerTypeRequired() const
 {
-    LayerType type = RenderPart::layerTypeRequired();
-    if (type != NoLayer)
-        return type;
-
-    return requiresAcceleratedCompositing() ? NormalLayer : NoLayer;
+    // This just use RenderPart::layerTypeRequired, because
RenderLayerCompositor
+    // doesn't loop through RenderEmbeddedObjects the way it does frames in
order
+    // to update the self painting bit on their RenderLayer.
+    if (requiresAcceleratedCompositing())
+        return NormalLayer;
+    return RenderPart::layerTypeRequired();
 }
 
 static String unavailablePluginReplacementText(Node* node,
RenderEmbeddedObject::PluginUnavailabilityReason pluginUnavailabilityReason)

[ojan, 2014-06-03 03:50:34]

The CQ bit was unchecked by ojan@chromium.org

[ojan, 2014-06-03 03:50:57]

This isn't ready to commit since the rebaselines are stale now. Will update
tomorrow and CQ.

[ojan, 2014-06-03 18:26:10]

Had to add back in RenderEmbeddedObject::layerTypeRequired. It was just removed
in https://codereview.chromium.org/309763002.

[ojan, 2014-06-03 18:26:17]

The CQ bit was checked by ojan@chromium.org

[commit-bot: I haz the power, 2014-06-03 18:28:50]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ojan@chromium.org/302083002/40001

[ojan, 2014-06-03 18:57:49]

The CQ bit was unchecked by ojan@chromium.org

[ojan, 2014-06-03 18:58:55]

Sigh...I'm conflicting with other NeedsRebaseline lines.

drpanke: FYI, this is the sort of thing that doing this off the trybots would
fix. I've now twice been unable to land this patch because of conflicting with
other NeedsRebaseline/NeedsManualRebaseline lines.

[Dirk Pranke, 2014-06-03 19:19:09]

On 2014/06/03 18:58:55, ojan wrote:
> drpanke: FYI, this is the sort of thing that doing this off the trybots would
> fix. I've now twice been unable to land this patch because of conflicting with
> other NeedsRebaseline/NeedsManualRebaseline lines.

Sure, there's no question that if fewer 'NeedsRebaseline' lines were getting
committed, we'd see fewer conflicts.

[ojan, 2014-06-03 22:48:22]

The CQ bit was checked by ojan@chromium.org

[commit-bot: I haz the power, 2014-06-03 22:48:47]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ojan@chromium.org/302083002/40001

[commit-bot: I haz the power, 2014-06-04 01:48:16]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  mac_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/10173)
  win_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_rel/builds/10494)

[ojan, 2014-06-04 01:49:21]

The CQ bit was unchecked by ojan@chromium.org

[ojan, 2014-06-04 01:50:07]

The CQ bit was checked by ojan@chromium.org

[commit-bot: I haz the power, 2014-06-04 01:50:30]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ojan@chromium.org/302083002/50001

[commit-bot: I haz the power, 2014-06-04 03:25:26]

Change committed as 175436