Rendering context information added to SharedQuadState

Rendering context information added to SharedQuadState

Context IDs given to DrawQuads to identify when a rendering context
changes, and whether or not a DrawQuad is in a 3D sorting context.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=278288

[Ian Vollick, 2014-06-13 02:25:13]

On 2014/06/13 00:35:09, troyhildebrandt wrote:

We actually get the context ids from blink and do a translation step in
web_layer_impl.cc to convert the ids to bools. Please plumb those through rather
than regenerating them with a counter. When you do, please also nuke
is_3d_sorted_ and base the return value of is_3d_sorted() on the context id.

[danakj, 2014-06-13 19:16:03]

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.cc
File cc/layers/layer.cc (right):

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.cc#newco...
cc/layers/layer.cc:758: void Layer::SetContextId(int id) {
can you give this name a bit more "context" (ha ha) in it's name, just the word
context doesn't give a reader an idea of what this id is. then please propagate
that more descriptive name throughout the rest of the CL.

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.cc#newco...
cc/layers/layer.cc:762: draw_properties_.context_id = id;
If we're setting this property with a Set method on Layer then it isn't a draw
property, either this Set method needs to go (and we set this inside
CalculateDrawProperties directly on the draw_props, or we make this a member of
the Layer class instead of draw_properties.

It looks to me like you want to remove this from draw_props and put it as a
member on Layer and LayerImpl instead (like is_3d_sorted_ was before)

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.h
File cc/layers/layer.h (right):

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.h#newcod...
cc/layers/layer.h:323: bool Is3dSorted() const { return
(draw_properties_.context_id != 0); }
no () needed in the function body

https://codereview.chromium.org/332873005/diff/20001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_impl.cc (right):

https://codereview.chromium.org/332873005/diff/20001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_impl.cc:655: root_layer->context_id());
I think you want a "not part of a context" id here, these quads are not part of
any layer they're the background behind all the layers.

[troyhildebrandt, 2014-06-13 22:56:36]

CC unittests and perftests were changed to reflect the lack of SetIs3dSorted().

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.cc
File cc/layers/layer.cc (right):

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.cc#newco...
cc/layers/layer.cc:758: void Layer::SetContextId(int id) {
On 2014/06/13 19:16:03, danakj wrote:
> can you give this name a bit more "context" (ha ha) in it's name, just the
word
> context doesn't give a reader an idea of what this id is. then please
propagate
> that more descriptive name throughout the rest of the CL.

Done.

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.cc#newco...
cc/layers/layer.cc:762: draw_properties_.context_id = id;
On 2014/06/13 19:16:03, danakj wrote:
> If we're setting this property with a Set method on Layer then it isn't a draw
> property, either this Set method needs to go (and we set this inside
> CalculateDrawProperties directly on the draw_props, or we make this a member
of
> the Layer class instead of draw_properties.
> 
> It looks to me like you want to remove this from draw_props and put it as a
> member on Layer and LayerImpl instead (like is_3d_sorted_ was before)

Done.

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.h
File cc/layers/layer.h (right):

https://codereview.chromium.org/332873005/diff/20001/cc/layers/layer.h#newcod...
cc/layers/layer.h:323: bool Is3dSorted() const { return
(draw_properties_.context_id != 0); }
On 2014/06/13 19:16:03, danakj wrote:
> no () needed in the function body

Done.

https://codereview.chromium.org/332873005/diff/20001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_impl.cc (right):

https://codereview.chromium.org/332873005/diff/20001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_impl.cc:655: root_layer->context_id());
On 2014/06/13 19:16:03, danakj wrote:
> I think you want a "not part of a context" id here, these quads are not part
of
> any layer they're the background behind all the layers.

Done.

[danakj, 2014-06-15 18:15:36]

Thanks! A few more comments.

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.cc
File cc/layers/layer.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.cc#newco...
cc/layers/layer.cc:902: layer->SetRenderingContextId(rendering_context_id_);
Don't need to do this twice

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.h
File cc/layers/layer.h (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.h#newcod...
cc/layers/layer.h:535: // The rendering context ID we get from Blink, plumbed
through WebLayerImpl
Can you say what its for rather than where it came from? In the ui compositor or
other embedders of cc this comment would be wrong.

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_impl_uni...
File cc/layers/layer_impl_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_impl_uni...
cc/layers/layer_impl_unittest.cc:158: // Set rendering context ID to 1 for 3D
sorting
Don't need these comments here. Were just setting some arbitrary value and
ensuring correct side effects.

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_position...
File cc/layers/layer_position_constraint_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_position...
cc/layers/layer_position_constraint_unittest.cc:26: bool is_3d_sorted) {
Nothing in this file even sets this to true. You could just remove the parameter
and the Set call

https://codereview.chromium.org/332873005/diff/40001/cc/output/renderer_pixel...
File cc/output/renderer_pixeltest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/output/renderer_pixel...
cc/output/renderer_pixeltest.cc:69: 0);
Can you please use a temp var to give these literals a name as we do for the
other arguments here. Same below.

https://codereview.chromium.org/332873005/diff/40001/cc/quads/draw_quad_unitt...
File cc/quads/draw_quad_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/quads/draw_quad_unitt...
cc/quads/draw_quad_unittest.cc:50: 0);
please give this a name with a temp var here and below also.

and can you use a value other than 0, as it would be much harder to accidentally
set the new thing to something other than 0. this just need to be any arbitrary
number that fits in the variable.

https://codereview.chromium.org/332873005/diff/40001/cc/test/layer_tree_json_...
File cc/test/layer_tree_json_parser.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/test/layer_tree_json_...
cc/test/layer_tree_json_parser.cc:134: new_layer->SetRenderingContextId(1);  //
A non-zero context ID = 3d sorted
can you make this comment a proper sentence (ie s/=/will make the layer be/)
with a period at the end. You can move the comment to its own line and wrap this
block with {} to include the comment and the Set call.

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_common_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_common_unittest.cc:3957: // Set rendering context ID to
1 for 3D sorting
comment nit: stick a period on the end, comments should be complete sentences

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_impl.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_impl.cc:655: 0);
can you give this literal a name by using a temporary variable, like we do for
opacity

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_pixeltest_filters.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_pixeltest_filters.cc:114:
background->SetRenderingContextId(1);  // For 3D sorting
comment periods

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_impl...
File cc/trees/layer_tree_impl_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_impl...
cc/trees/layer_tree_impl_unittest.cc:986: root->SetRenderingContextId(1);  //
For 3D sorting
If every call of this method needs a comment explaining it is for 3d sorting,
maybe we should improve the method name. How about Set3dSortingContextId()?

I know this matches the WebLayerImpl name, but it seems like we should improve
that name too then.

[troyhildebrandt, 2014-06-17 00:29:43]

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.cc
File cc/layers/layer.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.cc#newco...
cc/layers/layer.cc:902: layer->SetRenderingContextId(rendering_context_id_);
On 2014/06/15 18:15:35, danakj wrote:
> Don't need to do this twice

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.h
File cc/layers/layer.h (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer.h#newcod...
cc/layers/layer.h:535: // The rendering context ID we get from Blink, plumbed
through WebLayerImpl
On 2014/06/15 18:15:35, danakj wrote:
> Can you say what its for rather than where it came from? In the ui compositor
or
> other embedders of cc this comment would be wrong. 

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_impl_uni...
File cc/layers/layer_impl_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_impl_uni...
cc/layers/layer_impl_unittest.cc:158: // Set rendering context ID to 1 for 3D
sorting
On 2014/06/15 18:15:35, danakj wrote:
> Don't need these comments here. Were just setting some arbitrary value and
> ensuring correct side effects. 

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_position...
File cc/layers/layer_position_constraint_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/layers/layer_position...
cc/layers/layer_position_constraint_unittest.cc:26: bool is_3d_sorted) {
On 2014/06/15 18:15:36, danakj wrote:
> Nothing in this file even sets this to true. You could just remove the
parameter
> and the Set call

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/output/renderer_pixel...
File cc/output/renderer_pixeltest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/output/renderer_pixel...
cc/output/renderer_pixeltest.cc:69: 0);
On 2014/06/15 18:15:36, danakj wrote:
> Can you please use a temp var to give these literals a name as we do for the
> other arguments here. Same below.  

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/quads/draw_quad_unitt...
File cc/quads/draw_quad_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/quads/draw_quad_unitt...
cc/quads/draw_quad_unittest.cc:50: 0);
On 2014/06/15 18:15:36, danakj wrote:
> please give this a name with a temp var here and below also.
> 
> and can you use a value other than 0, as it would be much harder to
accidentally
> set the new thing to something other than 0. this just need to be any
arbitrary
> number that fits in the variable.

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/test/layer_tree_json_...
File cc/test/layer_tree_json_parser.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/test/layer_tree_json_...
cc/test/layer_tree_json_parser.cc:134: new_layer->SetRenderingContextId(1);  //
A non-zero context ID = 3d sorted
On 2014/06/15 18:15:36, danakj wrote:
> can you make this comment a proper sentence (ie s/=/will make the layer be/)
> with a period at the end. You can move the comment to its own line and wrap
this
> block with {} to include the comment and the Set call.

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_common_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_common_unittest.cc:3957: // Set rendering context ID to
1 for 3D sorting
On 2014/06/15 18:15:36, danakj wrote:
> comment nit: stick a period on the end, comments should be complete sentences

Removed the comment instead, since the method name is now more descriptive.

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_impl.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_impl.cc:655: 0);
On 2014/06/15 18:15:36, danakj wrote:
> can you give this literal a name by using a temporary variable, like we do for
> opacity

Done.

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
File cc/trees/layer_tree_host_pixeltest_filters.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_host...
cc/trees/layer_tree_host_pixeltest_filters.cc:114:
background->SetRenderingContextId(1);  // For 3D sorting
On 2014/06/15 18:15:36, danakj wrote:
> comment periods

Removed the comments instead.

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_impl...
File cc/trees/layer_tree_impl_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/40001/cc/trees/layer_tree_impl...
cc/trees/layer_tree_impl_unittest.cc:986: root->SetRenderingContextId(1);  //
For 3D sorting
On 2014/06/15 18:15:36, danakj wrote:
> If every call of this method needs a comment explaining it is for 3d sorting,
> maybe we should improve the method name. How about Set3dSortingContextId()?
> 
> I know this matches the WebLayerImpl name, but it seems like we should improve
> that name too then.

Changed to Set3dSortingContextId for Layer/LayerImpl, but changing it in
WebLayerImpl requires making changes to WebLayer.h, GraphicsLayer.cpp/h, and
CompositedLayerMapping.cpp and I'm not sure if we want to change these at this
point.

[enne (OOO), 2014-06-17 00:46:04]

(Could you also word wrap your patch description at 72 characters?)

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer.h
File cc/layers/layer.h (right):

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer.h#newcod...
cc/layers/layer.h:535: // The rendering context ID, necessary for properly
sorting 3D layers.
This comment is also not that helpful.  Maybe something like what I suggested
for LayerImpl would be helpful here.

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer_impl.h
File cc/layers/layer_impl.h (right):

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer_impl.h#n...
cc/layers/layer_impl.h:668: // The rendering context ID we get from Blink,
plumbed through WebLayerImpl
This doesn't necessarily come from Blink, just from the cc embedder.  It's also
not a very useful comment.

What about "Layers that share a sorting context id will be sorted together in 3d
space.  0 is a special value that means this layer will not be sorted and will
be drawn in paint order."

[troyhildebrandt, 2014-06-17 16:33:20]

Made the comment changes and wrapped the patch description at 72 characters.

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer.h
File cc/layers/layer.h (right):

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer.h#newcod...
cc/layers/layer.h:535: // The rendering context ID, necessary for properly
sorting 3D layers.
On 2014/06/17 00:46:04, enne wrote:
> This comment is also not that helpful.  Maybe something like what I suggested
> for LayerImpl would be helpful here.

Done.

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer_impl.h
File cc/layers/layer_impl.h (right):

https://codereview.chromium.org/332873005/diff/60001/cc/layers/layer_impl.h#n...
cc/layers/layer_impl.h:668: // The rendering context ID we get from Blink,
plumbed through WebLayerImpl
On 2014/06/17 00:46:04, enne wrote:
> This doesn't necessarily come from Blink, just from the cc embedder.  It's
also
> not a very useful comment.
> 
> What about "Layers that share a sorting context id will be sorted together in
3d
> space.  0 is a special value that means this layer will not be sorted and will
> be drawn in paint order."

Done.

[danakj, 2014-06-17 17:48:52]

Thanks for all the fixes, LGTM

[troyhildebrandt, 2014-06-17 18:03:14]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-17 18:05:23]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/80001

[troyhildebrandt, 2014-06-17 18:43:14]

The CQ bit was unchecked by [email protected]

[troyhildebrandt, 2014-06-17 21:31:00]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-17 21:32:06]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/100001

[troyhildebrandt, 2014-06-17 21:49:34]

cevans: content/common/cc_messages.h OWNERS
One addition to the SharedQuadState IPC messages.

[troyhildebrandt, 2014-06-17 21:53:21]

The CQ bit was unchecked by [email protected]

[troyhildebrandt, 2014-06-17 21:53:34]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-17 21:55:09]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/100001

[troyhildebrandt, 2014-06-17 22:10:05]

The CQ bit was unchecked by [email protected]

[troyhildebrandt, 2014-06-17 23:06:40]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-17 23:08:43]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/120001

[commit-bot: I haz the power, 2014-06-18 05:18:08]

The CQ bit was unchecked by [email protected]

[commit-bot: I haz the power, 2014-06-18 05:18:09]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/chromium_presubmit/bu...)

[danakj, 2014-06-18 15:29:46]

https://codereview.chromium.org/332873005/diff/120001/content/common/cc_messa...
File content/common/cc_messages_unittest.cc (right):

https://codereview.chromium.org/332873005/diff/120001/content/common/cc_messa...
content/common/cc_messages_unittest.cc:65: EXPECT_EQ(a->blend_mode,
b->blend_mode);
Compare the new field here please

[kenrb, 2014-06-18 15:38:45]

ipc lgtm

[troyhildebrandt, 2014-06-18 17:06:57]

On 2014/06/18 at 15:29:46, danakj wrote:
>
https://codereview.chromium.org/332873005/diff/120001/content/common/cc_messa...
> File content/common/cc_messages_unittest.cc (right):
> 
>
https://codereview.chromium.org/332873005/diff/120001/content/common/cc_messa...
> content/common/cc_messages_unittest.cc:65: EXPECT_EQ(a->blend_mode,
b->blend_mode);
> Compare the new field here please

Done.

[danakj, 2014-06-18 17:08:13]

Thanks! LGTMagain.

+piman for content/ OWNERS

[piman, 2014-06-18 21:08:13]

LGTM for content/

[troyhildebrandt, 2014-06-18 21:17:20]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-18 21:18:47]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/140001

[troyhildebrandt, 2014-06-18 21:55:26]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-18 21:56:58]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/160001

[troyhildebrandt, 2014-06-18 22:14:53]

The CQ bit was unchecked by [email protected]

[troyhildebrandt, 2014-06-18 23:20:41]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-06-18 23:23:49]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/[email protected]/332873005/160001

[commit-bot: I haz the power, 2014-06-19 06:49:33]

Change committed as 278288