Add a gpu.SwapBuffers synthetic delay to PostSubBuffer.

Rename gpu.SwapBuffers to gpu.PresentingFrame and add it to PostSubBuffer.

Depending on the GL implementation, PostSubBuffer may be called instead of
SwapBuffers.  Rename gpu.SwapBuffers to a more generic gpu.PresentingFrame
and update all the tests that use the delay.

BUG=359662
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=265936

[Geoff Lang, 2014-04-17 17:23:17]

PTAL

[Zhenyao Mo, 2014-04-17 20:45:47]

https://codereview.chromium.org/240073008/diff/1/gpu/command_buffer/service/g...
File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):

https://codereview.chromium.org/240073008/diff/1/gpu/command_buffer/service/g...
gpu/command_buffer/service/gles2_cmd_decoder.cc:7551:
TRACE_EVENT_SYNTHETIC_DELAY("gpu.SwapBuffers");
I feel you should use a different name from "gpu.SwapBuffers", like
"gpu.PostSubBuffer".  You may have to set the time for the new name in several
places, but in the long run, this avoids any unexpected side-effects /confusion
in the future.

[Geoff Lang, 2014-04-17 20:47:22]

On 2014/04/17 20:45:47, Zhenyao Mo wrote:
>
https://codereview.chromium.org/240073008/diff/1/gpu/command_buffer/service/g...
> File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):
> 
>
https://codereview.chromium.org/240073008/diff/1/gpu/command_buffer/service/g...
> gpu/command_buffer/service/gles2_cmd_decoder.cc:7551:
> TRACE_EVENT_SYNTHETIC_DELAY("gpu.SwapBuffers");
> I feel you should use a different name from "gpu.SwapBuffers", like
> "gpu.PostSubBuffer".  You may have to set the time for the new name in several
> places, but in the long run, this avoids any unexpected side-effects
/confusion
> in the future.

We talked about this a bit in the bug and the drawback is that we'd have to
update a lot of tests to put delays in both gpu.PostSubBuffer and
gpu.SwapBuffers since it's not possible to know which will be called.

[Zhenyao Mo, 2014-04-17 20:53:44]

On 2014/04/17 20:47:22, Geoff Lang wrote:
> On 2014/04/17 20:45:47, Zhenyao Mo wrote:
> >
>
https://codereview.chromium.org/240073008/diff/1/gpu/command_buffer/service/g...
> > File gpu/command_buffer/service/gles2_cmd_decoder.cc (right):
> > 
> >
>
https://codereview.chromium.org/240073008/diff/1/gpu/command_buffer/service/g...
> > gpu/command_buffer/service/gles2_cmd_decoder.cc:7551:
> > TRACE_EVENT_SYNTHETIC_DELAY("gpu.SwapBuffers");
> > I feel you should use a different name from "gpu.SwapBuffers", like
> > "gpu.PostSubBuffer".  You may have to set the time for the new name in
several
> > places, but in the long run, this avoids any unexpected side-effects
> /confusion
> > in the future.
> 
> We talked about this a bit in the bug and the drawback is that we'd have to
> update a lot of tests to put delays in both gpu.PostSubBuffer and
> gpu.SwapBuffers since it's not possible to know which will be called.

I still think a different name is better. For example, people may not remember
you have a SwapBuffers delay in PostSubBuffers unless they come across this
code.  I don't know what scenario, but in general we should avoid such hacking
and take the heavier way to set time for the new name for all tests.

[Zhenyao Mo, 2014-04-17 20:54:28]

Don't want to hold your progress. Let's seek a third party (piman) to decide.

[piman, 2014-04-18 00:56:42]

On 2014/04/17 20:54:28, Zhenyao Mo wrote:
> Don't want to hold your progress. Let's seek a third party (piman) to decide.

Maybe the best is to rename the event "PresentingFrame" or something like this
to make it orthogonal to the API used?

[Geoff Lang, 2014-04-18 00:58:08]

On 2014/04/18 00:56:42, piman wrote:
> On 2014/04/17 20:54:28, Zhenyao Mo wrote:
> > Don't want to hold your progress. Let's seek a third party (piman) to
decide.
> 
> Maybe the best is to rename the event "PresentingFrame" or something like this
> to make it orthogonal to the API used?

That works for me.

[Geoff Lang, 2014-04-22 17:40:26]

PTAL, updated the delay name to gpu.PresentingFrame and updated the tests.

[piman, 2014-04-22 17:45:01]

LGTM, thanks!

[Zhenyao Mo, 2014-04-22 17:56:17]

LGTM

[nduca, 2014-04-23 23:10:19]

lgtm

[Geoff Lang, 2014-04-23 23:18:48]

The CQ bit was checked by geofflang@chromium.org

[commit-bot: I haz the power, 2014-04-23 23:19:16]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/geofflang@chromium.org/240073008/20001

[commit-bot: I haz the power, 2014-04-24 00:12:06]

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

[commit-bot: I haz the power, 2014-04-24 00:12:06]

Try jobs failed on following builders:
  tryserver.chromium on chromium_presubmit

[Geoff Lang, 2014-04-24 13:41:21]

The CQ bit was checked by geofflang@chromium.org

[commit-bot: I haz the power, 2014-04-24 14:15:11]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/geofflang@chromium.org/240073008/20001

[commit-bot: I haz the power, 2014-04-24 16:40:29]

Change committed as 265936