Added Tough Texture upload page set to smoothness benchmark.

Added Tough Texture upload page set to smoothness benchmark.

Tough texture uploads need to be measured in our perf tests as well,
these tests will give us a sense of when we have a regression in how
we handle texture uploads.

I have also written a couple of pages that continuously upload textures
using WebGL. These tests upload textures of various sizes.

R=dtu@chromium.org, nednguyen@chromium.org
BUG=462078
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect

Committed: https://crrev.com/9a7070b677a106eec194520562d32f0af071a408
Cr-Commit-Position: refs/heads/master@{#331616}

[David Yen, 2015-05-26 16:22:09]

dyen@chromium.org changed reviewers:
+ nednguyen@google.com
- nednguyen@chromium.org

[David Yen, 2015-05-26 16:23:35]

PTAL. One thing I ran into was for some reason I did not seem to be getting
results for the large/extra_large texture uploads. Is there an issue where
results don't get outputted if the tests run too slowly?

[nednguyen, 2015-05-26 16:54:17]

On 2015/05/26 16:23:35, David Yen wrote:
> PTAL. One thing I ran into was for some reason I did not seem to be getting
> results for the large/extra_large texture uploads. Is there an issue where
> results don't get outputted if the tests run too slowly?

Maybe it has something to do with the results filtering mechanism? I recently
realize that it's easy to shoot oneself in the foot with it.

[David Yen, 2015-05-26 16:57:18]

On 2015/05/26 16:54:17, nednguyen wrote:
> On 2015/05/26 16:23:35, David Yen wrote:
> > PTAL. One thing I ran into was for some reason I did not seem to be getting
> > results for the large/extra_large texture uploads. Is there an issue where
> > results don't get outputted if the tests run too slowly?
> 
> Maybe it has something to do with the results filtering mechanism? I recently
> realize that it's easy to shoot oneself in the foot with it.

It's possible but the small/medium pages work and I couldn't spot any
differences between small/medium pages vs large/extra large uploads pages. The
only other difference I can think of is the large/extra large pages run much
slower. With my totally unmeasured estimate the large probably runs around 5fps.

[David Yen, 2015-05-26 17:47:41]

On 2015/05/26 16:57:18, David Yen wrote:
> On 2015/05/26 16:54:17, nednguyen wrote:
> > On 2015/05/26 16:23:35, David Yen wrote:
> > > PTAL. One thing I ran into was for some reason I did not seem to be
getting
> > > results for the large/extra_large texture uploads. Is there an issue where
> > > results don't get outputted if the tests run too slowly?
> > 
> > Maybe it has something to do with the results filtering mechanism? I
recently
> > realize that it's easy to shoot oneself in the foot with it.
> 
> It's possible but the small/medium pages work and I couldn't spot any
> differences between small/medium pages vs large/extra large uploads pages. The
> only other difference I can think of is the large/extra large pages run much
> slower. With my totally unmeasured estimate the large probably runs around
5fps.

Thanks for the help, fixed the issue where results were not being properly
calculated.

[nednguyen, 2015-05-26 17:50:46]

https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
File tools/perf/page_sets/tough_texture_upload_cases.py (right):

https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
tools/perf/page_sets/tough_texture_upload_cases.py:19: action_runner.Wait(10)
Since these are pages that we control, can we do a more proper wait, i.e:
WaitForAnimatioToStop(...)?

[David Yen, 2015-05-26 17:53:44]

https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
File tools/perf/page_sets/tough_texture_upload_cases.py (right):

https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
tools/perf/page_sets/tough_texture_upload_cases.py:19: action_runner.Wait(10)
On 2015/05/26 17:50:46, nednguyen wrote:
> Since these are pages that we control, can we do a more proper wait, i.e:
> WaitForAnimatioToStop(...)?

All of these pages just loop an animation infinitely so it should be fine to
just control how many seconds we want to record here.

[nednguyen, 2015-05-26 18:22:56]

On 2015/05/26 17:53:44, David Yen wrote:
>
https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
> File tools/perf/page_sets/tough_texture_upload_cases.py (right):
> 
>
https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
> tools/perf/page_sets/tough_texture_upload_cases.py:19: action_runner.Wait(10)
> On 2015/05/26 17:50:46, nednguyen wrote:
> > Since these are pages that we control, can we do a more proper wait, i.e:
> > WaitForAnimatioToStop(...)?
> 
> All of these pages just loop an animation infinitely so it should be fine to
> just control how many seconds we want to record here.

lgtm if the numbers look right on your local machine.

[David Yen, 2015-05-26 18:28:58]

On 2015/05/26 18:22:56, nednguyen wrote:
> On 2015/05/26 17:53:44, David Yen wrote:
> >
>
https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
> > File tools/perf/page_sets/tough_texture_upload_cases.py (right):
> > 
> >
>
https://codereview.chromium.org/1159733003/diff/60001/tools/perf/page_sets/to...
> > tools/perf/page_sets/tough_texture_upload_cases.py:19:
action_runner.Wait(10)
> > On 2015/05/26 17:50:46, nednguyen wrote:
> > > Since these are pages that we control, can we do a more proper wait, i.e:
> > > WaitForAnimatioToStop(...)?
> > 
> > All of these pages just loop an animation infinitely so it should be fine to
> > just control how many seconds we want to record here.
> 
> lgtm if the numbers look right on your local machine.

Numbers looks reasonable, on my nexus 5 I'm getting:

  small texture uploads - 60fps
  medium texture uploads - 30fps
  large texture uploads - 7fsp
  extra large texture uploads - 2fps

[David Yen, 2015-05-26 18:29:06]

The CQ bit was checked by dyen@chromium.org

[commit-bot: I haz the power, 2015-05-26 18:29:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1159733003/60001

[commit-bot: I haz the power, 2015-05-26 18:40:07]

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

[commit-bot: I haz the power, 2015-05-26 18:40:08]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
  linux_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisec...)
  mac_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_perf_bisect/...)

[David Yen, 2015-05-27 15:26:14]

The CQ bit was checked by dyen@chromium.org

[commit-bot: I haz the power, 2015-05-27 15:26:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1159733003/60001

[commit-bot: I haz the power, 2015-05-27 15:33:51]

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

[commit-bot: I haz the power, 2015-05-27 15:33:52]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[David Yen, 2015-05-27 16:12:49]

The CQ bit was checked by dyen@chromium.org

[commit-bot: I haz the power, 2015-05-27 16:13:16]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1159733003/60001

[commit-bot: I haz the power, 2015-05-27 16:21:20]

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

[commit-bot: I haz the power, 2015-05-27 16:21:22]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[nednguyen, 2015-05-27 16:41:37]

On 2015/05/27 16:21:22, I haz the power - commit-bot wrote:
> Try jobs failed on following builders:
>   chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

The presubmit check failure is legit, fyi

[David Yen, 2015-05-27 16:42:37]

On 2015/05/27 16:41:37, nednguyen wrote:
> On 2015/05/27 16:21:22, I haz the power - commit-bot wrote:
> > Try jobs failed on following builders:
> >   chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
> >
>
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
> 
> The presubmit check failure is legit, fyi

Yeah I just fixed it, I thought it was complaining about the perf bisect job
again.

[David Yen, 2015-05-27 16:56:00]

The CQ bit was checked by dyen@chromium.org

[David Yen, 2015-05-27 16:56:00]

The patchset sent to the CQ was uploaded after l-g-t-m from nednguyen@google.com
Link to the patchset: https://codereview.chromium.org/1159733003/#ps80001
(title: "Added copyright")

[commit-bot: I haz the power, 2015-05-27 16:56:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1159733003/80001

[commit-bot: I haz the power, 2015-05-27 19:01:05]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-05-27 19:02:05]

Patchset 5 (id:??) landed as
https://crrev.com/9a7070b677a106eec194520562d32f0af071a408
Cr-Commit-Position: refs/heads/master@{#331616}