Disable GPU acceleration on 2D canvas when readbacks are needed

Disable GPU acceleration on 2D canvas when readbacks are needed

We are already disabling GPU acceleration to avoid readbacks caused
by calls to getImageData. This change applies the same principle
to canvas-to-canvas drawImage calls in order to avoid probable
future readbacks.

BUG=652126, 651517, 650116, 642539, 640144
Committed: https://crrev.com/2b0d65c9311d7e87784e90d60bfc68efb5f03555
Cr-Commit-Position: refs/heads/master@{#422848}

[Justin Novosad, 2016-10-03 20:40:19]

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

[commit-bot: I haz the power, 2016-10-03 20:40:50]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Justin Novosad, 2016-10-03 20:40:51]

junov@chromium.org changed reviewers:
+ xidachen@chromium.org, xlai@chromium.org

[Justin Novosad, 2016-10-03 20:40:52]

PTAL

[commit-bot: I haz the power, 2016-10-03 21:53:01]

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

[commit-bot: I haz the power, 2016-10-03 21:53:02]

Dry run: Try jobs failed on following builders:
  linux_chromium_rel_ng on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[xidachen, 2016-10-04 00:48:58]

lgtm

[Justin Novosad, 2016-10-04 14:01:39]

The CQ bit was checked by junov@chromium.org

[Justin Novosad, 2016-10-04 14:01:39]

The patchset sent to the CQ was uploaded after l-g-t-m from
xidachen@chromium.org
Link to the patchset: https://codereview.chromium.org/2388293002/#ps20001
(title: "fixed tests")

[commit-bot: I haz the power, 2016-10-04 14:01:52]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[xidachen, 2016-10-04 14:03:43]

On 2016/10/04 14:01:52, commit-bot: I haz the power wrote:
> CQ is trying da patch. Follow status at
>  
>
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

From the trybots in PS#1, I think you need to rebaseline this test:
virtual/gpu/fast/canvas/canvas-imageSmoothingEnabled-patterns.html

There are about 5 pixels difference...

[Justin Novosad, 2016-10-04 14:25:12]

On 2016/10/04 14:03:43, xidachen wrote:
> On 2016/10/04 14:01:52, commit-bot: I haz the power wrote:
> > CQ is trying da patch. Follow status at
> >  
> >
>
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
> 
> From the trybots in PS#1, I think you need to rebaseline this test:
> virtual/gpu/fast/canvas/canvas-imageSmoothingEnabled-patterns.html
> 
> There are about 5 pixels difference...

Nope, no rebaseline. Instead I added a check for the canvas2dFixedRenderingMode
flag in order to disable the heuristic in layout tests.  This way we retain test
coverage for the GPU path.

[commit-bot: I haz the power, 2016-10-04 17:36:50]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2016-10-04 17:38:11]

Description was changed from

==========
Disable GPU acceleration on 2D canvas when readbacks are needed

We are already disabling GPU acceleration to avoid readbacks caused
by calls to getImageData. This change applies the same principle
to canvas-to-canvas drawImage calls in order to avoid probable
future readbacks.

BUG=652126, 651517, 650116, 642539, 640144
==========

to

==========
Disable GPU acceleration on 2D canvas when readbacks are needed

We are already disabling GPU acceleration to avoid readbacks caused
by calls to getImageData. This change applies the same principle
to canvas-to-canvas drawImage calls in order to avoid probable
future readbacks.

BUG=652126, 651517, 650116, 642539, 640144

Committed: https://crrev.com/2b0d65c9311d7e87784e90d60bfc68efb5f03555
Cr-Commit-Position: refs/heads/master@{#422848}
==========

[commit-bot: I haz the power, 2016-10-04 17:38:12]

Patchset 2 (id:??) landed as
https://crrev.com/2b0d65c9311d7e87784e90d60bfc68efb5f03555
Cr-Commit-Position: refs/heads/master@{#422848}

[Stephen White, 2016-10-04 18:40:13]

senorblanco@chromium.org changed reviewers:
+ senorblanco@chromium.org

[Stephen White, 2016-10-04 18:40:14]

Post-landing nits

https://codereview.chromium.org/2388293002/diff/20001/third_party/WebKit/Sour...
File
third_party/WebKit/Source/platform/graphics/ExpensiveCanvasHeuristicParameters.h
(right):

https://codereview.chromium.org/2388293002/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/platform/graphics/ExpensiveCanvasHeuristicParameters.h:72:
// from the gpu is necessary. This option cause the source canvas to
Nit: cause -> causes

https://codereview.chromium.org/2388293002/diff/20001/third_party/WebKit/Sour...
third_party/WebKit/Source/platform/graphics/ExpensiveCanvasHeuristicParameters.h:74:
// prevent future canvas to canvas draws from requiring a readback.
uNit: "canvas to canvas" -> "canvas-to-canvas"