Mac ÜC: Enable back-pressure for the browser compositor

Mac ÜC: Enable back-pressure for the browser compositor

Delay acknowledging the cc::OutputSurface's swap completion
until the surface is displayed.

BUG=314190
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=276629

[ccameron, 2014-06-12 01:01:46]

This turns out to be simpler than the lock scheme. The ui::Compositor implements
backpressure, so the fact that the cc::SingleThreadProxy doesn't isn't an issue.

I went with the approach of putting the Mac-specific bits inside an #ifdef --
let me know if you'd prefer that they be present on all platforms.

[piman, 2014-06-12 01:09:51]

lgtm

[danakj, 2014-06-12 01:16:52]

LGTM2 \o/ with 1 question

https://codereview.chromium.org/334593002/diff/20001/cc/output/output_surface.cc
File cc/output/output_surface.cc (right):

https://codereview.chromium.org/334593002/diff/20001/cc/output/output_surface...
cc/output/output_surface.cc:345:
base::Bind(&OutputSurface::DoOnSwapBuffersComplete,
Does virtual method in a posttask not do what you'd expect it to do, or why do
you need this indirection through DoOnSwapBuffersComplete, and not just bind
OnSwapBuffersComplete here directly?

[ccameron, 2014-06-12 01:28:26]

Thanks!!

https://codereview.chromium.org/334593002/diff/20001/cc/output/output_surface.cc
File cc/output/output_surface.cc (right):

https://codereview.chromium.org/334593002/diff/20001/cc/output/output_surface...
cc/output/output_surface.cc:345:
base::Bind(&OutputSurface::DoOnSwapBuffersComplete,
On 2014/06/12 01:16:51, danakj wrote:
> Does virtual method in a posttask not do what you'd expect it to do, or why do
> you need this indirection through DoOnSwapBuffersComplete, and not just bind
> OnSwapBuffersComplete here directly?

I was under the impression that a virtual method sent to posttask doesn't work
(well, it calls the precise method listed, not the derived class method).

But... I just tested this empirically, and it totally works! Huh! I wonder how
that's implemented...

[ccameron, 2014-06-12 07:02:22]

The CQ bit was checked by ccameron@chromium.org

[commit-bot: I haz the power, 2014-06-12 07:06:59]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ccameron@chromium.org/334593002/40001

[commit-bot: I haz the power, 2014-06-12 11:34:47]

Change committed as 276629

[piman, 2014-06-12 16:50:28]

On Wed, Jun 11, 2014 at 6:28 PM, <ccameron@chromium.org> wrote:

> Thanks!!
>
>
>
> https://codereview.chromium.org/334593002/diff/20001/cc/
> output/output_surface.cc
> File cc/output/output_surface.cc (right):
>
> https://codereview.chromium.org/334593002/diff/20001/cc/
> output/output_surface.cc#newcode345
> cc/output/output_surface.cc:345:
> base::Bind(&OutputSurface::DoOnSwapBuffersComplete,
> On 2014/06/12 01:16:51, danakj wrote:
>
>> Does virtual method in a posttask not do what you'd expect it to do,
>>
> or why do
>
>> you need this indirection through DoOnSwapBuffersComplete, and not
>>
> just bind
>
>> OnSwapBuffersComplete here directly?
>>
>
> I was under the impression that a virtual method sent to posttask
> doesn't work (well, it calls the precise method listed, not the derived
> class method).
>
> But... I just tested this empirically, and it totally works! Huh! I
> wonder how that's implemented...
>

Dereferencing a pointer to virtual function actually calls the derived
class method, not the base class.

Antoine


> https://codereview.chromium.org/334593002/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.