gpu: Add CHROMIUM_schedule_ca_layer

gpu: Add CHROMIUM_schedule_ca_layer

This is the variant on CHROMIUM_schedule_overlay_plane that accepts all
of the various arguments that are needed to represent most pages
entirely as CALayers.

BUG=533690
Committed: https://crrev.com/7494ac9b96578a7e6348283607626a63f5294e7b
Cr-Commit-Position: refs/heads/master@{#358808}

[ccameron, 2015-11-08 06:23:50]

ccameron@chromium.org changed reviewers:
+ bajones@chromium.org

[ccameron, 2015-11-08 06:23:51]

This is the plumbing that I was threatening about earlier.

[bajones, 2015-11-09 19:42:23]

command buffer bits LGTM w/ a couple of nits, but I have no idea about the
CALayer stuff.

https://codereview.chromium.org/1409193007/diff/40001/gpu/GLES2/extensions/CH...
File gpu/GLES2/extensions/CHROMIUM/CHROMIUM_schedule_ca_layer.txt (right):

https://codereview.chromium.org/1409193007/diff/40001/gpu/GLES2/extensions/CH...
gpu/GLES2/extensions/CHROMIUM/CHROMIUM_schedule_ca_layer.txt:3:
CHROMIUM_schedule_overlay_plane
Typo? Should this be CHROMIUM_schedule_ca_layer

https://codereview.chromium.org/1409193007/diff/40001/gpu/GLES2/extensions/CH...
gpu/GLES2/extensions/CHROMIUM/CHROMIUM_schedule_ca_layer.txt:7:
GL_CHROMIUM_schedule_overlay_plane
Ditto. GL_CHROMIUM_schedule_ca_layer

https://codereview.chromium.org/1409193007/diff/40001/gpu/command_buffer/serv...
File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):

https://codereview.chromium.org/1409193007/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:9471: gfx::Transform
transform(mem[6], mem[10], mem[14], mem[18], mem[7], mem[11],
Nit: I had to stare at this for a bit to understand it. Could you put the line
breaks after every 4 values to make it easier to read at a glance? Thanks!

[ccameron, 2015-11-09 20:04:07]

Thanks!

Adding ben@ to stamp mojo/* auto-gen code.

Adding avi@ for ui/*

https://codereview.chromium.org/1409193007/diff/40001/gpu/GLES2/extensions/CH...
File gpu/GLES2/extensions/CHROMIUM/CHROMIUM_schedule_ca_layer.txt (right):

https://codereview.chromium.org/1409193007/diff/40001/gpu/GLES2/extensions/CH...
gpu/GLES2/extensions/CHROMIUM/CHROMIUM_schedule_ca_layer.txt:3:
CHROMIUM_schedule_overlay_plane
On 2015/11/09 19:42:22, bajones wrote:
> Typo? Should this be CHROMIUM_schedule_ca_layer

Done.

https://codereview.chromium.org/1409193007/diff/40001/gpu/GLES2/extensions/CH...
gpu/GLES2/extensions/CHROMIUM/CHROMIUM_schedule_ca_layer.txt:7:
GL_CHROMIUM_schedule_overlay_plane
On 2015/11/09 19:42:22, bajones wrote:
> Ditto. GL_CHROMIUM_schedule_ca_layer

Done.

https://codereview.chromium.org/1409193007/diff/40001/gpu/command_buffer/serv...
File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):

https://codereview.chromium.org/1409193007/diff/40001/gpu/command_buffer/serv...
gpu/command_buffer/service/gles2_cmd_decoder.cc:9471: gfx::Transform
transform(mem[6], mem[10], mem[14], mem[18], mem[7], mem[11],
On 2015/11/09 19:42:22, bajones wrote:
> Nit: I had to stare at this for a bit to understand it. Could you put the line
> breaks after every 4 values to make it easier to read at a glance? Thanks!

git cl format strikes again!

Moved it back to 4x4.

[ccameron, 2015-11-10 02:10:16]

ccameron@chromium.org changed reviewers:
+ avi@chromium.org, ben@chromium.org

[ccameron, 2015-11-10 02:10:16]

(this time for real)

Adding ben@ to stamp mojo/* auto-gen code.

Adding avi@ for ui/* and CoreAnimation

[Avi (use Gerrit), 2015-11-10 02:25:58]

This code looks entirely plausible to me, but I have zero CA experience and
can't give this the review it deserves. I suspect you know the graphics
reviewers better than I do.

https://codereview.chromium.org/1409193007/diff/60001/content/common/gpu/imag...
File content/common/gpu/image_transport_surface_overlay_mac.mm (right):

https://codereview.chromium.org/1409193007/diff/60001/content/common/gpu/imag...
content/common/gpu/image_transport_surface_overlay_mac.mm:8: 
Why is this include separate from all the other ones below?

https://codereview.chromium.org/1409193007/diff/60001/ui/gl/gl_surface.h
File ui/gl/gl_surface.h (right):

https://codereview.chromium.org/1409193007/diff/60001/ui/gl/gl_surface.h#newc...
ui/gl/gl_surface.h:30: class VSyncProvider;
alphabetize?

[ccameron, 2015-11-10 03:05:06]

Thanks -- updated.

Unfortunately the only other Mac guy left who knows CoreAnimation is tapted@,
and he's out for a few weeks -- I can have him TBR for when he gets back.

https://codereview.chromium.org/1409193007/diff/60001/content/common/gpu/imag...
File content/common/gpu/image_transport_surface_overlay_mac.mm (right):

https://codereview.chromium.org/1409193007/diff/60001/content/common/gpu/imag...
content/common/gpu/image_transport_surface_overlay_mac.mm:8: 
On 2015/11/10 02:25:58, Avi wrote:
> Why is this include separate from all the other ones below?

Done.

https://codereview.chromium.org/1409193007/diff/60001/ui/gl/gl_surface.h
File ui/gl/gl_surface.h (right):

https://codereview.chromium.org/1409193007/diff/60001/ui/gl/gl_surface.h#newc...
ui/gl/gl_surface.h:30: class VSyncProvider;
On 2015/11/10 02:25:58, Avi wrote:
> alphabetize?

Done.

[Avi (use Gerrit), 2015-11-10 03:30:38]

I thought it was going to be some massive CA patch, but it's just making simple
calls on CALayer, so I'm ok taking it. LGTM with a minor nit.

https://codereview.chromium.org/1409193007/diff/80001/content/common/gpu/imag...
File content/common/gpu/image_transport_surface_overlay_mac.mm (right):

https://codereview.chromium.org/1409193007/diff/80001/content/common/gpu/imag...
content/common/gpu/image_transport_surface_overlay_mac.mm:172: [ca_layer
setAnchorPoint:CGPointMake(0, 0)];
CGZeroPoint?

https://codereview.chromium.org/1409193007/diff/80001/content/common/gpu/imag...
content/common/gpu/image_transport_surface_overlay_mac.mm:206: [ca_layer
setBackgroundColor:CGColorGetConstantColor(kCGColorClear)];
You do CGColorGetConstantColor(kCGColorClear) repeatedly (here and above). Is it
worth globally caching it, or is that premature optimization?

[Ben Goodger (Google), 2015-11-10 04:40:13]

mojo lgtm

[ccameron, 2015-11-10 07:15:43]

Thanks!

https://codereview.chromium.org/1409193007/diff/80001/content/common/gpu/imag...
File content/common/gpu/image_transport_surface_overlay_mac.mm (right):

https://codereview.chromium.org/1409193007/diff/80001/content/common/gpu/imag...
content/common/gpu/image_transport_surface_overlay_mac.mm:172: [ca_layer
setAnchorPoint:CGPointMake(0, 0)];
On 2015/11/10 03:30:38, Avi wrote:
> CGZeroPoint?

Done.

https://codereview.chromium.org/1409193007/diff/80001/content/common/gpu/imag...
content/common/gpu/image_transport_surface_overlay_mac.mm:206: [ca_layer
setBackgroundColor:CGColorGetConstantColor(kCGColorClear)];
On 2015/11/10 03:30:38, Avi wrote:
> You do CGColorGetConstantColor(kCGColorClear) repeatedly (here and above). Is
it
> worth globally caching it, or is that premature optimization?

Yes, there's this, and I also CGColorSpaceCreateWithName later on. I haven't put
this into Instruments to find the hotspots yet -- I'm holding off on
optimizations until then.

[ccameron, 2015-11-10 07:15:51]

The CQ bit was checked by ccameron@chromium.org

[ccameron, 2015-11-10 07:15:52]

The patchset sent to the CQ was uploaded after l-g-t-m from
bajones@chromium.org, avi@chromium.org, ben@chromium.org
Link to the patchset: https://codereview.chromium.org/1409193007/#ps100001
(title: "Incorporate review feedback")

[commit-bot: I haz the power, 2015-11-10 07:16:19]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1409193007/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1409193007/100001

[commit-bot: I haz the power, 2015-11-10 08:46:19]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-11-10 08:47:20]

Patchset 6 (id:??) landed as
https://crrev.com/7494ac9b96578a7e6348283607626a63f5294e7b
Cr-Commit-Position: refs/heads/master@{#358808}