Add better support for RGBA_4444 textures.

Add better support for RGBA_4444 textures.

Add support for RGBA_4444 textures to one/zero-copy. Also
adds command line flags that allow RGBA_4444 textures to
be enabled/disabled for testing purposes.

BUG=511284
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/cfd85eeea491f720f2dd2fcf310fe634b237e601
Cr-Commit-Position: refs/heads/master@{#344585}

[reveman, 2015-08-10 23:19:00]

reveman@chromium.org changed reviewers:
+ danakj@chromium.org, vmpstr@chromium.org

[reveman, 2015-08-10 23:19:01]

This fixes RGBA_4444 support and adds command line flags that make it possible
to exercise the code on any platform.

This depends on https://codereview.chromium.org/1230203007

https://codereview.chromium.org/1248253003/diff/40001/cc/raster/tile_task_wor...
File cc/raster/tile_task_worker_pool.cc (right):

https://codereview.chromium.org/1248253003/diff/40001/cc/raster/tile_task_wor...
cc/raster/tile_task_worker_pool.cc:228: canvas_bitmap_rect, scale);
Note: this change is needed to make the code below work with partial updates. It
basically disables partial updates when we have to convert pixels. I think we
can improve this but I prefer to keep it simple until texture compression
patches have landed.

[danakj, 2015-08-10 23:35:55]

https://codereview.chromium.org/1248253003/diff/40001/cc/raster/tile_task_wor...
File cc/raster/tile_task_worker_pool.cc (right):

https://codereview.chromium.org/1248253003/diff/40001/cc/raster/tile_task_wor...
cc/raster/tile_task_worker_pool.cc:228: canvas_bitmap_rect, scale);
On 2015/08/10 23:19:01, reveman (OOO Aug 10 - Aug 11) wrote:
> Note: this change is needed to make the code below work with partial updates.
It
> basically disables partial updates when we have to convert pixels. I think we
> can improve this but I prefer to keep it simple until texture compression
> patches have landed. 

Can you file a bug and point to it in a TODO saying that this should be
canvas_playback_rect to get partial updates and why it's not etc?

https://codereview.chromium.org/1248253003/diff/40001/ui/compositor/composito...
File ui/compositor/compositor_switches.h (right):

https://codereview.chromium.org/1248253003/diff/40001/ui/compositor/composito...
ui/compositor/compositor_switches.h:16: COMPOSITOR_EXPORT extern const char
kUIEnableRGBA4444Textures[];
This isn't used

[reveman, 2015-08-10 23:49:42]

https://codereview.chromium.org/1248253003/diff/40001/cc/raster/tile_task_wor...
File cc/raster/tile_task_worker_pool.cc (right):

https://codereview.chromium.org/1248253003/diff/40001/cc/raster/tile_task_wor...
cc/raster/tile_task_worker_pool.cc:228: canvas_bitmap_rect, scale);
On 2015/08/10 at 23:35:54, danakj wrote:
> On 2015/08/10 23:19:01, reveman (OOO Aug 10 - Aug 11) wrote:
> > Note: this change is needed to make the code below work with partial
updates. It
> > basically disables partial updates when we have to convert pixels. I think
we
> > can improve this but I prefer to keep it simple until texture compression
> > patches have landed. 
> 
> Can you file a bug and point to it in a TODO saying that this should be
canvas_playback_rect to get partial updates and why it's not etc?

Added TODO.

https://codereview.chromium.org/1248253003/diff/40001/ui/compositor/composito...
File ui/compositor/compositor_switches.h (right):

https://codereview.chromium.org/1248253003/diff/40001/ui/compositor/composito...
ui/compositor/compositor_switches.h:16: COMPOSITOR_EXPORT extern const char
kUIEnableRGBA4444Textures[];
On 2015/08/10 at 23:35:54, danakj wrote:
> This isn't used

Bad rebase. Fixed in latest patch.

[danakj, 2015-08-11 17:37:42]

LGTM % vmpstr

[vmpstr, 2015-08-11 17:46:17]

lgtm

https://codereview.chromium.org/1248253003/diff/60001/cc/raster/tile_task_wor...
File cc/raster/tile_task_worker_pool.cc (right):

https://codereview.chromium.org/1248253003/diff/60001/cc/raster/tile_task_wor...
cc/raster/tile_task_worker_pool.cc:233: TRACE_EVENT0("cc",
"TileTaskWorkerPool::PlaybackToMemory::ConvertPixels");
nit: can you add a trace at the top of the function as well, so that we can tell
what's PlaybackToMemory taking and what portion of that is ConvertPixels

[reveman, 2015-08-11 18:11:29]

https://codereview.chromium.org/1248253003/diff/60001/cc/raster/tile_task_wor...
File cc/raster/tile_task_worker_pool.cc (right):

https://codereview.chromium.org/1248253003/diff/60001/cc/raster/tile_task_wor...
cc/raster/tile_task_worker_pool.cc:233: TRACE_EVENT0("cc",
"TileTaskWorkerPool::PlaybackToMemory::ConvertPixels");
On 2015/08/11 at 17:46:17, vmpstr wrote:
> nit: can you add a trace at the top of the function as well, so that we can
tell what's PlaybackToMemory taking and what portion of that is ConvertPixels

Done.

[reveman, 2015-08-19 16:27:20]

reveman@chromium.org changed reviewers:
+ piman@chromium.org

[reveman, 2015-08-19 16:27:20]

+piman for content/ and gpu/

[piman, 2015-08-19 21:13:25]

https://codereview.chromium.org/1248253003/diff/80001/content/renderer/gpu/re...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/1248253003/diff/80001/content/renderer/gpu/re...
content/renderer/gpu/render_widget_compositor.cc:468:
!cmd->HasSwitch(switches::kDisableRGBA4444Textures);
These flags won't be forwarded to the renderer unless they're added to
RenderProcessHostImpl::PropagateBrowserCommandLineToRenderer

[reveman, 2015-08-20 10:44:58]

https://codereview.chromium.org/1248253003/diff/80001/content/renderer/gpu/re...
File content/renderer/gpu/render_widget_compositor.cc (right):

https://codereview.chromium.org/1248253003/diff/80001/content/renderer/gpu/re...
content/renderer/gpu/render_widget_compositor.cc:468:
!cmd->HasSwitch(switches::kDisableRGBA4444Textures);
On 2015/08/19 at 21:13:25, piman (slow to review) wrote:
> These flags won't be forwarded to the renderer unless they're added to
RenderProcessHostImpl::PropagateBrowserCommandLineToRenderer

Done. Thanks.

[reveman, 2015-08-20 14:35:14]

The CQ bit was checked by reveman@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-08-20 14:35:40]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1248253003/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1248253003/100001

[commit-bot: I haz the power, 2015-08-20 16:14:31]

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

[commit-bot: I haz the power, 2015-08-20 16:14:32]

Dry run: This issue passed the CQ dry run.

[piman, 2015-08-20 20:55:36]

lgtm

[reveman, 2015-08-20 21:01:57]

The CQ bit was checked by reveman@chromium.org

[reveman, 2015-08-20 21:01:58]

The patchset sent to the CQ was uploaded after l-g-t-m from danakj@chromium.org,
vmpstr@chromium.org
Link to the patchset: https://codereview.chromium.org/1248253003/#ps100001
(title: "rebase and forward flags")

[commit-bot: I haz the power, 2015-08-20 21:03:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1248253003/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1248253003/100001

[commit-bot: I haz the power, 2015-08-20 21:14:08]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-08-20 21:14:44]

Patchset 6 (id:??) landed as
https://crrev.com/cfd85eeea491f720f2dd2fcf310fe634b237e601
Cr-Commit-Position: refs/heads/master@{#344585}