Mac: Fix bugs in Sierra color fix

Mac: Fix bugs in Sierra color fix

The previous fix uncondiontally replaced populated the solid color
IOSurface and updated the contents_rect for all layers on Sierra,
instead of just doing it for the ones without IOSurfaces.

Fix this bug, parameterize whether or not we are using this workaround,
and add tests for both code paths.

BUG=633805
Committed: https://crrev.com/496c26b5cc3983f40b65729d8fa8bbd4a9951c71
Cr-Commit-Position: refs/heads/master@{#422281}

[ccameron, 2016-09-30 22:14:18]

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

[ccameron, 2016-09-30 22:14:18]

Arguably, we should just throw out our use of solid color layers on all OS
versions, and unconditionally use the solid color IOSurface path. WDYT?

https://codereview.chromium.org/2387653004/diff/1/ui/accelerated_widget_mac/c...
File ui/accelerated_widget_mac/ca_renderer_layer_tree.mm (left):

https://codereview.chromium.org/2387653004/diff/1/ui/accelerated_widget_mac/c...
ui/accelerated_widget_mac/ca_renderer_layer_tree.mm:369: if
(base::mac::IsAtLeastOS10_12()) {
I was testing this on 10.11, so I screwed up this conditional and made us hit
the below lines unconditionally.

[erikchen, 2016-09-30 22:21:35]

lgtm

On 2016/09/30 22:14:18, ccameron wrote:
> Arguably, we should just throw out our use of solid color layers on all OS
> versions, and unconditionally use the solid color IOSurface path. WDYT?
Assuming there are no performance penalties, less logic = less likely to make a
mistake. Can you throw up a 1-line CL and use telemetry try-jobs?

> 
>
https://codereview.chromium.org/2387653004/diff/1/ui/accelerated_widget_mac/c...
> File ui/accelerated_widget_mac/ca_renderer_layer_tree.mm (left):
> 
>
https://codereview.chromium.org/2387653004/diff/1/ui/accelerated_widget_mac/c...
> ui/accelerated_widget_mac/ca_renderer_layer_tree.mm:369: if
> (base::mac::IsAtLeastOS10_12()) {
> I was testing this on 10.11, so I screwed up this conditional and made us hit
> the below lines unconditionally.

[ccameron, 2016-09-30 22:42:12]

On 2016/09/30 22:21:35, erikchen wrote:
> lgtm
> 
> On 2016/09/30 22:14:18, ccameron wrote:
> > Arguably, we should just throw out our use of solid color layers on all OS
> > versions, and unconditionally use the solid color IOSurface path. WDYT?
> Assuming there are no performance penalties, less logic = less likely to make
a
> mistake. Can you throw up a 1-line CL and use telemetry try-jobs?

Good call, will-do.

[ccameron, 2016-09-30 22:42:15]

The CQ bit was checked by ccameron@chromium.org

[commit-bot: I haz the power, 2016-09-30 22:42:41]

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

[commit-bot: I haz the power, 2016-10-01 02:10:14]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-10-01 02:13:45]

Description was changed from

==========
Mac: Fix bugs in Sierra color fix

The previous fix uncondiontally replaced populated the solid color
IOSurface and updated the contents_rect for all layers on Sierra,
instead of just doing it for the ones without IOSurfaces.

Fix this bug, parameterize whether or not we are using this workaround,
and add tests for both code paths.

BUG=633805
==========

to

==========
Mac: Fix bugs in Sierra color fix

The previous fix uncondiontally replaced populated the solid color
IOSurface and updated the contents_rect for all layers on Sierra,
instead of just doing it for the ones without IOSurfaces.

Fix this bug, parameterize whether or not we are using this workaround,
and add tests for both code paths.

BUG=633805

Committed: https://crrev.com/496c26b5cc3983f40b65729d8fa8bbd4a9951c71
Cr-Commit-Position: refs/heads/master@{#422281}
==========

[commit-bot: I haz the power, 2016-10-01 02:13:46]

Patchset 1 (id:??) landed as
https://crrev.com/496c26b5cc3983f40b65729d8fa8bbd4a9951c71
Cr-Commit-Position: refs/heads/master@{#422281}