cc: fix for Ganesh in MSAA mode.

cc: fix for Ganesh in MSAA mode.

When rendering in Ganesh with MSAA enabled, request that skia flush each
canvas after drawing a tile. This will cause the MSAA render targets
to be resolved to the shared texture which CC subsequently draws.

BUG=

Committed: https://crrev.com/41a787b1699c647e57794465bc271da2cd4b9643
Cr-Commit-Position: refs/heads/master@{#313761}

[Stephen White, 2015-01-29 15:16:11]

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

[Stephen White, 2015-01-29 15:16:18]

Dana: PTAL. Thanks!

[danakj, 2015-01-29 16:57:10]

On 2015/01/29 15:16:18, Stephen White wrote:
> Dana: PTAL. Thanks!

Do you think we should be pixel testing this stuff?

[Stephen White, 2015-01-29 18:02:39]

On 2015/01/29 16:57:10, danakj wrote:
> On 2015/01/29 15:16:18, Stephen White wrote:
> > Dana: PTAL. Thanks!
> 
> Do you think we should be pixel testing this stuff?

I'd say no, just because the MSAA results tend to be very vendor-specific (e.g.,
Mesa could completely change their algorithm, and still be spec-compliant).

[danakj, 2015-01-29 18:11:30]

On 2015/01/29 18:02:39, Stephen White wrote:
> On 2015/01/29 16:57:10, danakj wrote:
> > On 2015/01/29 15:16:18, Stephen White wrote:
> > > Dana: PTAL. Thanks!
> > 
> > Do you think we should be pixel testing this stuff?
> 
> I'd say no, just because the MSAA results tend to be very vendor-specific
(e.g.,
> Mesa could completely change their algorithm, and still be spec-compliant).

LGTM. I think we need some kind of story for testing GPU raster though.

[Stephen White, 2015-01-29 18:13:22]

On 2015/01/29 18:11:30, danakj wrote:
> On 2015/01/29 18:02:39, Stephen White wrote:
> > On 2015/01/29 16:57:10, danakj wrote:
> > > On 2015/01/29 15:16:18, Stephen White wrote:
> > > > Dana: PTAL. Thanks!
> > > 
> > > Do you think we should be pixel testing this stuff?
> > 
> > I'd say no, just because the MSAA results tend to be very vendor-specific
> (e.g.,
> > Mesa could completely change their algorithm, and still be spec-compliant).
> 
> LGTM. I think we need some kind of story for testing GPU raster though.

I've logged a Chrome bug for it: crbug.com/453490

[Stephen White, 2015-01-29 18:13:32]

The CQ bit was checked by senorblanco@chromium.org

[commit-bot: I haz the power, 2015-01-29 18:14:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/885883002/1

[commit-bot: I haz the power, 2015-01-29 19:41:39]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-01-29 19:43:55]

Patchset 1 (id:??) landed as
https://crrev.com/41a787b1699c647e57794465bc271da2cd4b9643
Cr-Commit-Position: refs/heads/master@{#313761}