Enable blink_perf.canvas on Android Perf bots

Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707, 592017
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel

Committed: https://crrev.com/7d700edf761d46092b24d53ee228ce0df6e7b054
Cr-Commit-Position: refs/heads/master@{#379686}

[dshwang, 2016-01-29 14:31:48]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
==========

[dshwang, 2016-01-29 14:32:31]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
==========

[dshwang, 2016-01-29 14:32:32]

dongseong.hwang@intel.com changed reviewers:
+ kbr@chromium.org, qyearsley@chromium.org

[dshwang, 2016-01-29 14:37:15]

kbr, could you review? I'll watch those bots and make sure the newly enabled
tests run cleanly once this lands.

The blocking issue was fixed; BUG=504773
So it may works well now.

This perf test was in trouble, especially Nexus 6 bot as usual (Adreno (TM) 420
:( ). I'll keep eyes on it.
https://codereview.chromium.org/1163573003#msg4

[Ken Russell (switch to Gerrit), 2016-01-30 02:14:05]

Thanks for re-visiting this. lgtm

[qyearsley, 2016-01-31 20:57:34]

The CQ bit was checked by qyearsley@chromium.org

[qyearsley, 2016-01-31 20:57:34]

lgtm

[commit-bot: I haz the power, 2016-01-31 20:57:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/1

[commit-bot: I haz the power, 2016-01-31 21:07:23]

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

[commit-bot: I haz the power, 2016-01-31 21:07:24]

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...)

[dshwang, 2016-02-01 10:53:19]

On 2016/01/31 21:07:24, commit-bot: I haz the power 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...)

android_nexus5_perf_bisect is added to trybot infra. Impressive.
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_...

However, external contributors don't have permission to read.
could you copy&paste the log here?

[qyearsley, 2016-02-01 18:17:29]

On 2016/02/01 10:53:19, dshwang wrote:
> On 2016/01/31 21:07:24, commit-bot: I haz the power 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...)
> 
> android_nexus5_perf_bisect is added to trybot infra. Impressive.
>
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_...
> 
> However, external contributors don't have permission to read.
> could you copy&paste the log here?

On 2016/02/01 at 10:53:19, dongseong.hwang wrote:
> On 2016/01/31 21:07:24, commit-bot: I haz the power 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...)
> 
> android_nexus5_perf_bisect is added to trybot infra. Impressive.
>
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_...
> 
> However, external contributors don't have permission to read.
> could you copy&paste the log here?

Hi -- the failure on mac_10_10_perf_bisect is unrelated to this patch. The
failure on android might be, but I'm not sure. It occurs when, during the
blink_perf.canvas benchmark run, the value reported for
'draw-video-to-hw-accelerated-canvas-2d' is an empty list.

Also, we'll need a Telemetry owner's stamp on this CL (that's what the
chromium_presubmit failure is). Owners listed in
https://chromium.googlesource.com/chromium/src/+/master/tools/perf/OWNERS.


Log contents for the failure on
android_nexus5_perf_bisect/builds/3239/steps/blink_perf.canvas:

Running 5 times
Ignoring warm-up run (112.59728529182503 runs/s)
124.00716761428808 runs/s
126.41266149217492 runs/s
126.692934335052 runs/s
127.08660316572724 runs/s
126.68330440731232 runs/s
Description: This bench test checks the speed on putting ImageData to
Canvas2D(1024x1024).

Time:
values 124.00716761428808, 126.41266149217492, 126.692934335052,
127.08660316572724, 126.68330440731232 runs/s
avg 126.17653420291091 runs/s
median 126.68330440731232 runs/s
stdev 1.2362865915913468 runs/s
min 124.00716761428808 runs/s
max 127.08660316572724 runs/s

[       OK ] putImageData.html (14443 ms)
[ RUN      ] draw-video-to-hw-accelerated-canvas-2d.html
Traceback (most recent call last):
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/third_party/catapult/telemetry/telemetry/internal/story_runner.py",
line 84, in _RunStoryAndProcessErrorIfNeeded
    state.RunStory(results)
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/third_party/catapult/telemetry/telemetry/page/shared_page_state.py",
line 329, in RunStory
    self._current_page, self._current_tab, results)
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/tools/perf/benchmarks/blink_perf.py",
line 108, in ValidateAndMeasurePage
    results.current_page, metric, units, values))
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/third_party/catapult/telemetry/telemetry/value/list_of_scalar_values.py",
line 82, in __init__
    assert len(values) > 0
AssertionError

[  FAILED  ] draw-video-to-hw-accelerated-canvas-2d.html (3240 ms)

Traceback (most recent call last):
  RunBenchmark at
third_party/catapult/telemetry/telemetry/internal/story_runner.py:302
    Run(pt, stories, finder_options, results, benchmark.max_failures)
  Run at third_party/catapult/telemetry/telemetry/internal/story_runner.py:219
    _RunStoryAndProcessErrorIfNeeded(story, results, state, test)
  _RunStoryAndProcessErrorIfNeeded at
third_party/catapult/telemetry/telemetry/internal/story_runner.py:84
    state.RunStory(results)
  RunStory at
third_party/catapult/telemetry/telemetry/page/shared_page_state.py:329
    self._current_page, self._current_tab, results)
  ValidateAndMeasurePage at tools/perf/benchmarks/blink_perf.py:108
    results.current_page, metric, units, values))
  __init__ at
third_party/catapult/telemetry/telemetry/value/list_of_scalar_values.py:82
    assert len(values) > 0
AssertionError

Locals:
  description            : None
  important              : True
  improvement_direction  : None
  name                   : 'draw-video-to-hw-accelerated-canvas-2d'
  none_value_reason      : None
  page                   : <telemetry.page.Page object at 0x7f6764a3a750>
  same_page_merge_policy : 'concatenate'
  std                    : None
  tir_label              : None
  units                  : u'runs/s'
  values                 : []

View generated profiler files online at
https://console.developers.google.com/m/cloudstorage/b/chrome-telemetry-outpu...
for page draw-video-to-hw-accelerated-canvas-2d.html
[  PASSED  ] 1 test.
[  FAILED  ] 1 test, listed below:
[  FAILED  ]  draw-video-to-hw-accelerated-canvas-2d.html

1 FAILED TEST

[dshwang, 2016-02-01 19:27:20]

Thank you for providing log.

I guess nexus 5 bot is also flaky, because my nexus 5 is fine with this CL. let
me try CQ again.

[dshwang, 2016-02-01 19:27:24]

The CQ bit was checked by dongseong.hwang@intel.com

[commit-bot: I haz the power, 2016-02-01 19:27:52]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/1

[commit-bot: I haz the power, 2016-02-01 19:36:12]

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

[commit-bot: I haz the power, 2016-02-01 19:36:13]

Try jobs failed on following builders:
  mac_10_10_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_b...)

[dtu, 2016-02-01 19:59:24]

dtu@chromium.org changed reviewers:
+ dtu@chromium.org

[dtu, 2016-02-01 19:59:25]

lgtm

[aiolos (Not reviewing), 2016-02-01 20:01:27]

aiolos@chromium.org changed reviewers:
+ aiolos@chromium.org

[aiolos (Not reviewing), 2016-02-01 20:01:28]

lgtm if it passes the android bisect.

Also, you may need to rebase since I think Quinten removed the xp disable in
another cl.

[qyearsley, 2016-02-03 18:39:15]

The CQ bit was checked by qyearsley@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-03 18:41:22]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/1

[commit-bot: I haz the power, 2016-02-03 18:47:16]

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

[commit-bot: I haz the power, 2016-02-03 18:47:17]

Dry run: Try jobs failed on following builders:
  android_clang_dbg_recipe on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_clan...)
  android_compile_dbg on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_comp...)
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
  linux_chromium_compile_dbg_32_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_compile_dbg_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
  ios_rel_device_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_rel_device_ni...)
  mac_chromium_compile_dbg_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_comp...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[qyearsley, 2016-02-03 18:56:03]

dongseong.hwang: This CL is waiting on a rebase now before it can be committed.

[dshwang, 2016-02-03 19:43:13]

On 2016/02/03 18:56:03, qyearsley wrote:
> dongseong.hwang: This CL is waiting on a rebase now before it can be
committed.

yep, rebased. let me try to land

[dshwang, 2016-02-03 19:43:18]

The CQ bit was checked by dongseong.hwang@intel.com

[dshwang, 2016-02-03 19:43:18]

The patchset sent to the CQ was uploaded after l-g-t-m from aiolos@chromium.org,
kbr@chromium.org, dtu@chromium.org, qyearsley@chromium.org
Link to the patchset: https://codereview.chromium.org/1644353002/#ps20001
(title: "rebase to ToT")

[commit-bot: I haz the power, 2016-02-03 19:44:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/20001

[commit-bot: I haz the power, 2016-02-03 21:48:08]

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

[commit-bot: I haz the power, 2016-02-03 21:48:09]

Try jobs failed on following builders:
  android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_TIMED_OUT, no build
URL)

[dshwang, 2016-02-04 21:10:35]

qyearsley, could you copy&paste logs again? Thank you. 
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_...

[aiolos (Not reviewing), 2016-02-04 21:27:48]

On 2016/02/04 21:10:35, dshwang wrote:
> qyearsley, could you copy&paste logs again? Thank you. 
>
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_...

It looks like the benchmark is trying to add values that don't exist. Not sure
why you aren't getting any results though.

Traceback (most recent call last):
  RunBenchmark at
third_party/catapult/telemetry/telemetry/internal/story_runner.py:302
    Run(pt, stories, finder_options, results, benchmark.max_failures)
  Run at third_party/catapult/telemetry/telemetry/internal/story_runner.py:219
    _RunStoryAndProcessErrorIfNeeded(story, results, state, test)
  _RunStoryAndProcessErrorIfNeeded at
third_party/catapult/telemetry/telemetry/internal/story_runner.py:84
    state.RunStory(results)
  RunStory at
third_party/catapult/telemetry/telemetry/page/shared_page_state.py:329
    self._current_page, self._current_tab, results)
  ValidateAndMeasurePage at tools/perf/benchmarks/blink_perf.py:109
    results.current_page, metric, units, values))
  __init__ at
third_party/catapult/telemetry/telemetry/value/list_of_scalar_values.py:82
    assert len(values) > 0
AssertionError

Locals:
  description            : None
  important              : True
  improvement_direction  : None
  name                   : 'upload-video-to-sub-texture'
  none_value_reason      : None
  page                   : <telemetry.page.Page object at 0x7f7f52c01410>
  same_page_merge_policy : 'concatenate'
  std                    : None
  tir_label              : None
  units                  : u'runs/s'
  values                 : []

(INFO) 2016-02-04 11:55:06,391 cloud_storage.Insert:270  Uploading
/tmp/tmp6bv9xC.png to
gs://chrome-telemetry-output/profiler-file-id_0-2016-02-04_11-55-0634117.png
View generated profiler files online at
https://console.developers.google.com/m/cloudstorage/b/chrome-telemetry-outpu...
for page upload-video-to-sub-texture.html
[  PASSED  ] 8 tests.
[  FAILED  ] 1 test, listed below:
[  FAILED  ]  upload-video-to-sub-texture.html

[dshwang, 2016-02-05 12:50:19]

Thank you for showing log.

I'm strongly suspecting the nexus 5 perf bot don't support mp4 codec decoding.
To make sure it, new patch set prints FATAL log when the test stop by unexpected
issue.
Let me kick dry bot.

[inactive_dshwang_plz_cc_intel, 2016-02-05 12:51:04]

The CQ bit was checked by dongseong.hwang@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-05 12:51:19]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/40001

[commit-bot: I haz the power, 2016-02-05 16:22:37]

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

[commit-bot: I haz the power, 2016-02-05 16:22:38]

Dry run: Try jobs failed on following builders:
  android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)

[qyearsley, 2016-02-05 17:43:53]

On 2016/02/05 at 16:22:38, commit-bot wrote:
> Dry run: Try jobs failed on following builders:
>   android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)

The log for the latest run now contains:

FATAL: test fails due to GPU issue. SecurityError: Failed to execute
'texSubImage2D' on 'WebGLRenderingContext': The video element contains
cross-origin data, and may not be loaded.

Which it didn't contain before. After that it contains the same message about
making an empty list of scalar values in benchmarks/blink_perf.py line 112.

[dshwang, 2016-02-05 19:39:08]

On 2016/02/05 17:43:53, qyearsley wrote:
> On 2016/02/05 at 16:22:38, commit-bot wrote:
> > Dry run: Try jobs failed on following builders:
> >   android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
>
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)
> 
> The log for the latest run now contains:
> 
> FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> cross-origin data, and may not be loaded.
> 
> Which it didn't contain before. After that it contains the same message about
> making an empty list of scalar values in benchmarks/blink_perf.py line 112.

Thank you again.

WebGL cannot draw video from file url. run_benchmark launchs local web server,
so WebGL should draw the video.
In my machine, run_benchmark is fine with both "release" and "android-chromium".

kbr, do you have idea why blink_perf has this issue, while webgl_conformance
tests are fine.
I could not find special command line in webgl_conformance.py
https://code.google.com/p/chromium/codesearch#chromium/src/content/test/gpu/g...

[inactive_dshwang_plz_cc_intel, 2016-02-05 20:00:27]

The CQ bit was checked by dongseong.hwang@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-05 20:01:08]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/60001

[commit-bot: I haz the power, 2016-02-06 00:14:31]

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

[commit-bot: I haz the power, 2016-02-06 00:14:32]

Dry run: Try jobs failed on following builders:
  android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_TIMED_OUT,
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)

[Ken Russell (switch to Gerrit), 2016-02-09 02:05:24]

On 2016/02/05 19:39:08, dshwang wrote:
> On 2016/02/05 17:43:53, qyearsley wrote:
> > On 2016/02/05 at 16:22:38, commit-bot wrote:
> > > Dry run: Try jobs failed on following builders:
> > >   android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
> >
>
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)
> > 
> > The log for the latest run now contains:
> > 
> > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> > 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> > cross-origin data, and may not be loaded.
> > 
> > Which it didn't contain before. After that it contains the same message
about
> > making an empty list of scalar values in benchmarks/blink_perf.py line 112.
> 
> Thank you again.
> 
> WebGL cannot draw video from file url. run_benchmark launchs local web server,
> so WebGL should draw the video.
> In my machine, run_benchmark is fine with both "release" and
"android-chromium".
> 
> kbr, do you have idea why blink_perf has this issue, while webgl_conformance
> tests are fine.
> I could not find special command line in webgl_conformance.py
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/test/gpu/g...

I don't know the blink_perf.canvas tests' code at all, but the following file
from the WebGL conformance suite might be helpful:
https://github.com/KhronosGroup/WebGL/blob/master/sdk/tests/js/webgl-test-uti...

Look for setupImageForCrossOriginTest and getLocalCrossOrigin.

[inactive_dshwang_plz_cc_intel, 2016-02-22 15:52:00]

The CQ bit was checked by dongseong.hwang@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-02-22 15:52:33]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/80001

[commit-bot: I haz the power, 2016-02-22 17:47:23]

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

[commit-bot: I haz the power, 2016-02-22 17:47:24]

Dry run: Try jobs failed on following builders:
  android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)

[dshwang, 2016-02-22 18:12:03]

sorry for delaying handling it. Let me restart.
I changed the video code according to webgl conformance tests, but
android_nexus5_perf_bisect still failed.

qyearsley@, could you copy&paste the log again? thank you in advance.

[qyearsley, 2016-02-23 00:23:30]

On 2016/02/22 at 18:12:03, dongseong.hwang wrote:
> sorry for delaying handling it. Let me restart.
> I changed the video code according to webgl conformance tests, but
android_nexus5_perf_bisect still failed.
> 
> qyearsley@, could you copy&paste the log again? thank you in advance.

Hmm, seems to be the same error as before:

FATAL: test fails due to GPU issue. SecurityError: Failed to execute
'texSubImage2D' on 'WebGLRenderingContext': The video element contains
cross-origin data, and may not be loaded.
Traceback (most recent call last):
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/third_party/catapult/telemetry/telemetry/internal/story_runner.py",
line 84, in _RunStoryAndProcessErrorIfNeeded
    state.RunStory(results)
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/third_party/catapult/telemetry/telemetry/page/shared_page_state.py",
line 325, in RunStory
    self._current_page, self._current_tab, results)
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/tools/perf/benchmarks/blink_perf.py",
line 112, in ValidateAndMeasurePage
    results.current_page, metric, units, values))
  File
"/b/build/slave/android_nexus5_perf_bisect/build/src/third_party/catapult/telemetry/telemetry/value/list_of_scalar_values.py",
line 82, in __init__
    assert len(values) > 0
AssertionError

[  FAILED  ] upload-video-to-sub-texture.html (6079 ms)

(Note, the failures on mac_chromium_10.10_rel_ng seem to be unrelated)

[dshwang, 2016-02-23 11:46:11]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[dshwang, 2016-02-23 12:00:09]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-23 12:00:36]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/100001

[commit-bot: I haz the power, 2016-02-23 12:51:18]

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

[commit-bot: I haz the power, 2016-02-23 12:51:19]

Dry run: Try jobs failed on following builders:
  linux_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  win_chromium_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[dshwang, 2016-02-23 15:20:20]

qyearsley, I added more printf to figure out why. Could you copy&paste one more
time? Thank you.

[qyearsley, 2016-02-23 17:53:46]

On 2016/02/23 15:20:20, dshwang wrote:
> qyearsley, I added more printf to figure out why. Could you copy&paste one
more
> time? Thank you.

Oh, this time it contains a slightly different error, and so the extra messages
you added were not printed!

FATAL: test fails due to GPU issue. SecurityError: Failed to execute
'getImageData' on 'CanvasRenderingContext2D': The canvas has been tainted by
cross-origin data.

What does it mean for a canvas (or video element) to be tainted with
cross-origin data? Which data is cross-origin here?

(The rest of the log appears the same as before, I think)

[dshwang, 2016-02-23 20:14:12]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-23 20:16:31]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/120001

[dshwang, 2016-02-23 20:18:31]

On 2016/02/23 17:53:46, qyearsley wrote:
> On 2016/02/23 15:20:20, dshwang wrote:
> > qyearsley, I added more printf to figure out why. Could you copy&paste one
> more
> > time? Thank you.
> 
> Oh, this time it contains a slightly different error, and so the extra
messages
> you added were not printed!
> 
> FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> 'getImageData' on 'CanvasRenderingContext2D': The canvas has been tainted by
> cross-origin data.
> 
> What does it mean for a canvas (or video element) to be tainted with
> cross-origin data? Which data is cross-origin here?
> 
> (The rest of the log appears the same as before, I think)

ouch, there are 3 video tests in trouble by same reason. 2 are webgl test and 1
is canvas 2d test.
upload-video-to-texture.html
upload-video-to-sub-texture.html
draw-video-to-hw-accelerated-canvas-2d.html

by change, this time test failed in canvas 2d test :(

I added the log code to canvas 2d code also. Could you copy&paste the nexus5 log
after the test failed again. thank you again.

[dshwang, 2016-02-23 20:21:25]

The CQ bit was unchecked by dongseong.hwang@intel.com

[dshwang, 2016-02-23 20:21:26]

The CQ bit was checked by dongseong.hwang@intel.com

[dshwang, 2016-02-23 20:21:26]

The patchset sent to the CQ was uploaded after l-g-t-m from aiolos@chromium.org,
kbr@chromium.org, dtu@chromium.org, qyearsley@chromium.org
Link to the patchset: https://codereview.chromium.org/1644353002/#ps120001
(title: "add log to canvas 2d also")

[commit-bot: I haz the power, 2016-02-23 20:22:24]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/120001

[commit-bot: I haz the power, 2016-02-23 20:28:03]

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

[commit-bot: I haz the power, 2016-02-23 20:28:04]

Try jobs failed on following builders:
  android_chromium_gn_compile_rel on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_chro...)
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)
  android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)

[dshwang, 2016-02-24 08:31:00]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-24 08:31:12]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/120001

[commit-bot: I haz the power, 2016-02-24 08:37:49]

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

[commit-bot: I haz the power, 2016-02-24 08:37:50]

Dry run: 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_chromium_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)
  mac_10_10_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_10_10_perf_b...)

[dshwang, 2016-02-24 09:33:08]

Patchset #7 (id:120001) has been deleted

[dshwang, 2016-02-24 09:35:22]

Patchset #7 (id:140001) has been deleted

[dshwang, 2016-02-24 09:36:44]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-24 09:37:14]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/160001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/160001

[commit-bot: I haz the power, 2016-02-24 09:55:28]

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

[commit-bot: I haz the power, 2016-02-24 09:55:30]

Dry run: Try jobs failed on following builders:
  android_nexus5_perf_bisect on tryserver.chromium.perf (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_p...)

[dshwang, 2016-02-24 13:19:49]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-24 13:20:32]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/160001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/160001

[commit-bot: I haz the power, 2016-02-24 14:38:01]

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

[commit-bot: I haz the power, 2016-02-24 14:38:03]

Dry run: Try jobs failed on following builders:
  win_chromium_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[dshwang, 2016-02-24 16:21:56]

qyearsley@, I added the log code to canvas 2d code also. Could you copy&paste
the nexus5 log?  thank you again.

There are 3 video tests in trouble by same reason. 2 are webgl test and 1
is canvas 2d test.
upload-video-to-texture.html
upload-video-to-sub-texture.html
draw-video-to-hw-accelerated-canvas-2d.html

by change, this time test failed in canvas 2d test :(

[qyearsley, 2016-02-24 18:29:05]

On 2016/02/24 at 16:21:56, dongseong.hwang wrote:
> qyearsley@, I added the log code to canvas 2d code also. Could you copy&paste
the nexus5 log?  thank you again.
> 
> There are 3 video tests in trouble by same reason. 2 are webgl test and 1
> is canvas 2d test.
> upload-video-to-texture.html
> upload-video-to-sub-texture.html
> draw-video-to-hw-accelerated-canvas-2d.html
> 
> by change, this time test failed in canvas 2d test :(

Ah, I see -- alright, now we have the added info in the log:

FATAL: test fails due to GPU issue. SecurityError: Failed to execute 'drawImage'
on 'CanvasRenderingContext2D': The video element contains cross-origin data at
http://127.0.0.1:43178/Canvas/resources/bear-1280x720.mp4, and may not be
loaded. hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
taintsCanvas:0

[Ken Russell (switch to Gerrit), 2016-02-24 18:58:48]

On 2016/02/24 18:29:05, qyearsley wrote:
> On 2016/02/24 at 16:21:56, dongseong.hwang wrote:
> > qyearsley@, I added the log code to canvas 2d code also. Could you
copy&paste
> the nexus5 log?  thank you again.
> > 
> > There are 3 video tests in trouble by same reason. 2 are webgl test and 1
> > is canvas 2d test.
> > upload-video-to-texture.html
> > upload-video-to-sub-texture.html
> > draw-video-to-hw-accelerated-canvas-2d.html
> > 
> > by change, this time test failed in canvas 2d test :(
> 
> Ah, I see -- alright, now we have the added info in the log:
> 
> FATAL: test fails due to GPU issue. SecurityError: Failed to execute
'drawImage'
> on 'CanvasRenderingContext2D': The video element contains cross-origin data at
> http://127.0.0.1:43178/Canvas/resources/bear-1280x720.mp4, and may not be
> loaded. hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
> taintsCanvas:0

What's the origin of the containing document? Is that in the log? Note that
http://localhost:[PORT]/ is a different domain than http://127.0.0.1:[PORT]/ .

[dshwang, 2016-02-24 20:26:10]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-24 20:27:17]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/180001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/180001

[dshwang, 2016-02-24 20:28:23]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[dshwang, 2016-02-24 20:28:23]

dongseong.hwang@intel.com changed reviewers:
+ qinmin@chromium.org

[commit-bot: I haz the power, 2016-02-24 20:31:42]

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

[commit-bot: I haz the power, 2016-02-24 20:31:44]

Dry run: Try jobs failed on following builders:
  android_arm64_dbg_recipe on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_arm6...)
  android_chromium_gn_compile_dbg on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_chro...)
  android_clang_dbg_recipe on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_clan...)
  android_compile_dbg on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/android_comp...)
  cast_shell_android on tryserver.chromium.android (JOB_FAILED,
https://build.chromium.org/p/tryserver.chromium.android/builders/cast_shell_a...)
  cast_shell_linux on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/cast_shell_linu...)
  chromeos_amd64-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_amd64-...)
  chromeos_daisy_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_daisy_...)
  chromeos_x86-generic_chromium_compile_only_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_x86-ge...)
  linux_chromium_asan_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_chromeos_compile_dbg_ng on tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_chromeos_ozone_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_clobber_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_compile_dbg_32_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_compile_dbg_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  linux_chromium_gn_chromeos_rel on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)
  ios_dbg_simulator_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
  ios_rel_device_ninja on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_rel_device_ni...)
  mac_chromium_compile_dbg_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_comp...)
  mac_chromium_gn_rel on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_gn_r...)
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)
  win8_chromium_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win8_chromium_ng/...)
  win_chromium_x64_rel_ng on tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_x64_...)

[dshwang, 2016-02-24 20:34:03]

Patchset #8 (id:180001) has been deleted

[dshwang, 2016-02-24 20:41:39]

On 2016/02/24 18:58:48, Ken Russell wrote:
> > Ah, I see -- alright, now we have the added info in the log:

Thank you!

> > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> 'drawImage'
> > on 'CanvasRenderingContext2D': The video element contains cross-origin data
at
> > http://127.0.0.1:43178/Canvas/resources/bear-1280x720.mp4, and may not be
> > loaded. hasWebMediaPlayer:1 hasSingleSecurityOrigin:0
didPassCORSAccessCheck:0
> > taintsCanvas:0
> 
> What's the origin of the containing document? Is that in the log? Note that
> http://localhost:[PORT]/ is a different domain than http://127.0.0.1:[PORT]/ .

document url should be
http://127.0.0.1:43178/Canvas/draw-video-to-hw-accelerated-canvas-2d.html
Let me check tomorrow to make sure it.

+qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
android_nexus5_perf_bisect. Do you have idea?
document url is
http://127.0.0.1:43178/Canvas/draw-video-to-hw-accelerated-canvas-2d.html
video url is http://127.0.0.1:43178/Canvas/resources/bear-1280x720.mp4
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...

[dshwang, 2016-02-25 18:18:18]

Patchset #8 (id:200001) has been deleted

[dshwang, 2016-02-25 18:23:05]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-25 18:27:06]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/220001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/220001

[commit-bot: I haz the power, 2016-02-25 19:22:51]

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

[commit-bot: I haz the power, 2016-02-25 19:22:53]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[dshwang, 2016-02-25 19:53:52]

On 2016/02/24 20:41:39, dshwang wrote:
> document url should be
> http://127.0.0.1:43178/Canvas/draw-video-to-hw-accelerated-canvas-2d.html
> Let me check tomorrow to make sure it.

qyearsley, could you copy&paste log from android_nexus5_perf_bisect one more
time? I added document url.
In addition, I added many LOG(ERROR) in webmediaplayer_android.cc. If the logs
is in stdio, please show me. It will be critical hint for qinmin.

> +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
> android_nexus5_perf_bisect. Do you have idea?
> document url is
> http://127.0.0.1:43178/Canvas/draw-video-to-hw-accelerated-canvas-2d.html
> video url is http://127.0.0.1:43178/Canvas/resources/bear-1280x720.mp4
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...

[qyearsley, 2016-02-25 22:48:26]

On 2016/02/25 19:53:52, dshwang wrote:
> On 2016/02/24 20:41:39, dshwang wrote:
> > document url should be
> > http://127.0.0.1:43178/Canvas/draw-video-to-hw-accelerated-canvas-2d.html
> > Let me check tomorrow to make sure it.
> 
> qyearsley, could you copy&paste log from android_nexus5_perf_bisect one more
> time? I added document url.
> In addition, I added many LOG(ERROR) in webmediaplayer_android.cc. If the logs
> is in stdio, please show me. It will be critical hint for qinmin.
> 
> > +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
> > android_nexus5_perf_bisect. Do you have idea?
> > document url is
> > http://127.0.0.1:43178/Canvas/draw-video-to-hw-accelerated-canvas-2d.html
> > video url is http://127.0.0.1:43178/Canvas/resources/bear-1280x720.mp4
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...

In the latest run of blink_perf.canvas on android_nexus5_perf_bisect, there are
no lines that have ERROR logs or any logs containing "hasSingleSecurityOrigin"
-- I believe that logs from  webmediaplayer_android.cc (or from anywhere in the
Chrome renderer or browser processes) are not printed in the Telemetry output.

[qyearsley, 2016-02-25 23:01:32]

> In the latest run of blink_perf.canvas on android_nexus5_perf_bisect, there
are
> no lines that have ERROR logs or any logs containing "hasSingleSecurityOrigin"
> -- I believe that logs from  webmediaplayer_android.cc (or from anywhere in
the
> Chrome renderer or browser processes) are not printed in the Telemetry output.

Is it feasible for anyone to run the test locally? Is a nexus 5 required or
might this reproduce on other devices? (If it's feasible to run it locally, then
you can also collect the browser log output by running run_benchmark with
--enable--browser-logging (and then the logs are output into a cryptically-named
OS-dependent temp location, e.g. /tmp/tmpHxdHaQ/chrome.log

[dshwang, 2016-02-26 06:48:53]

On 2016/02/25 23:01:32, qyearsley wrote:
> > In the latest run of blink_perf.canvas on android_nexus5_perf_bisect, there
> are
> > no lines that have ERROR logs or any logs containing
"hasSingleSecurityOrigin"
> > -- I believe that logs from  webmediaplayer_android.cc (or from anywhere in
> the
> > Chrome renderer or browser processes) are not printed in the Telemetry
output.
> 
> Is it feasible for anyone to run the test locally? Is a nexus 5 required or
> might this reproduce on other devices? (If it's feasible to run it locally,
then
> you can also collect the browser log output by running run_benchmark with
> --enable--browser-logging (and then the logs are output into a
cryptically-named
> OS-dependent temp location, e.g. /tmp/tmpHxdHaQ/chrome.log

So why does
https://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_...
fails?

I have Nexus5 but I cannot reproduce it. My Nexus5 just pass the perf tests :(
Let me kick dry run one more time.

[dshwang, 2016-02-26 06:49:21]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-26 06:49:36]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/220001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/220001

[commit-bot: I haz the power, 2016-02-26 06:55:31]

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

[commit-bot: I haz the power, 2016-02-26 06:55:34]

Dry run: Try jobs failed on following builders:
  mac_chromium_rel_ng on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_rel_...)

[dshwang, 2016-02-26 14:23:54]

On 2016/02/25 23:01:32, qyearsley wrote:
> > In the latest run of blink_perf.canvas on android_nexus5_perf_bisect, there
> are
> > no lines that have ERROR logs or any logs containing
"hasSingleSecurityOrigin"
> > -- I believe that logs from  webmediaplayer_android.cc (or from anywhere in
> the
> > Chrome renderer or browser processes) are not printed in the Telemetry
output.
> 
> Is it feasible for anyone to run the test locally? Is a nexus 5 required or
> might this reproduce on other devices? (If it's feasible to run it locally,
then
> you can also collect the browser log output by running run_benchmark with
> --enable--browser-logging (and then the logs are output into a
cryptically-named
> OS-dependent temp location, e.g. /tmp/tmpHxdHaQ/chrome.log

I kicked dryrun again. Could you check please?

[qyearsley, 2016-02-26 16:49:22]

On 2016/02/26 at 14:23:54, dongseong.hwang wrote:
> On 2016/02/25 23:01:32, qyearsley wrote:
> > > In the latest run of blink_perf.canvas on android_nexus5_perf_bisect,
there
> > are
> > > no lines that have ERROR logs or any logs containing
"hasSingleSecurityOrigin"
> > > -- I believe that logs from  webmediaplayer_android.cc (or from anywhere
in
> > the
> > > Chrome renderer or browser processes) are not printed in the Telemetry
output.
> > 
> > Is it feasible for anyone to run the test locally? Is a nexus 5 required or
> > might this reproduce on other devices? (If it's feasible to run it locally,
then
> > you can also collect the browser log output by running run_benchmark with
> > --enable--browser-logging (and then the logs are output into a
cryptically-named
> > OS-dependent temp location, e.g. /tmp/tmpHxdHaQ/chrome.log
> 
> I kicked dryrun again. Could you check please?

Last try job run had a patch error, which usually means you need to rebase and
upload again.

[dshwang, 2016-02-26 18:24:10]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[commit-bot: I haz the power, 2016-02-26 18:25:03]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/240001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/240001

[commit-bot: I haz the power, 2016-02-26 19:22:10]

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

[commit-bot: I haz the power, 2016-02-26 19:22:11]

Dry run: Try jobs failed on following builders:
  linux_chromium_clobber_rel_ng on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_...)

[dshwang, 2016-02-27 11:39:46]

On 2016/02/26 16:49:22, qyearsley wrote:
> Last try job run had a patch error, which usually means you need to rebase and
> upload again.

oh, sorry. I rebased it to ToT. Could you check it again?

[dshwang, 2016-03-02 11:00:53]

On 2016/02/27 11:39:46, dshwang wrote:
> On 2016/02/26 16:49:22, qyearsley wrote:
> > Last try job run had a patch error, which usually means you need to rebase
and
> > upload again.
> 
> oh, sorry. I rebased it to ToT. Could you check it again?

qyearsley, could you have a chance to check  android_nexus5_perf_bisect  again?
Thank you.

[qyearsley, 2016-03-02 17:29:35]

On 2016/03/02 11:00:53, dshwang wrote:
> On 2016/02/27 11:39:46, dshwang wrote:
> > On 2016/02/26 16:49:22, qyearsley wrote:
> > > Last try job run had a patch error, which usually means you need to rebase
> and
> > > upload again.
> > 
> > oh, sorry. I rebased it to ToT. Could you check it again?
> 
> qyearsley, could you have a chance to check  android_nexus5_perf_bisect 
again?
> Thank you.

From latest run on android_nexus5_perf_bisect:

FATAL: test fails due to GPU issue. SecurityError: Failed to execute
'texSubImage2D' on 'WebGLRenderingContext': The video element contains
cross-origin data at http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
and may not be loaded.
documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
taintsCanvas:0

[qyearsley, 2016-03-02 17:29:36]

On 2016/03/02 11:00:53, dshwang wrote:
> On 2016/02/27 11:39:46, dshwang wrote:
> > On 2016/02/26 16:49:22, qyearsley wrote:
> > > Last try job run had a patch error, which usually means you need to rebase
> and
> > > upload again.
> > 
> > oh, sorry. I rebased it to ToT. Could you check it again?
> 
> qyearsley, could you have a chance to check  android_nexus5_perf_bisect 
again?
> Thank you.

From latest run on android_nexus5_perf_bisect:

FATAL: test fails due to GPU issue. SecurityError: Failed to execute
'texSubImage2D' on 'WebGLRenderingContext': The video element contains
cross-origin data at http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
and may not be loaded.
documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
taintsCanvas:0

[dshwang, 2016-03-02 17:56:23]

On 2016/03/02 17:29:36, qyearsley wrote:
> On 2016/03/02 11:00:53, dshwang wrote:
> > On 2016/02/27 11:39:46, dshwang wrote:
> > > On 2016/02/26 16:49:22, qyearsley wrote:
> > > > Last try job run had a patch error, which usually means you need to
rebase
> > and
> > > > upload again.
> > > 
> > > oh, sorry. I rebased it to ToT. Could you check it again?
> > 
> > qyearsley, could you have a chance to check  android_nexus5_perf_bisect 
> again?
> > Thank you.
> 
> From latest run on android_nexus5_perf_bisect:
> 
> FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> cross-origin data at
http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> and may not be loaded.
> documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
> taintsCanvas:0

Thank you. Is there any LOG(ERROR) from webmediaplayer_android.cc ?

[qyearsley, 2016-03-02 18:38:06]

On 2016/03/02 17:56:23, dshwang wrote:
> On 2016/03/02 17:29:36, qyearsley wrote:
> > On 2016/03/02 11:00:53, dshwang wrote:
> > > On 2016/02/27 11:39:46, dshwang wrote:
> > > > On 2016/02/26 16:49:22, qyearsley wrote:
> > > > > Last try job run had a patch error, which usually means you need to
> rebase
> > > and
> > > > > upload again.
> > > > 
> > > > oh, sorry. I rebased it to ToT. Could you check it again?
> > > 
> > > qyearsley, could you have a chance to check  android_nexus5_perf_bisect 
> > again?
> > > Thank you.
> > 
> > From latest run on android_nexus5_perf_bisect:
> > 
> > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> > 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> > cross-origin data at
> http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > and may not be loaded.
> > documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
> > taintsCanvas:0
> 
> Thank you. Is there any LOG(ERROR) from webmediaplayer_android.cc ?

Nope. As mentioned earlier, I believe that logs from  webmediaplayer_android.cc
(or from anywhere in
the Chrome renderer or browser processes) are not printed in the Telemetry
output. But if someone can run the test locally, one can pass
--enable--browser-logging to run_benchmark, and then the logs are output into a
temp file.

[dshwang, 2016-03-02 18:50:55]

On 2016/03/02 18:38:06, qyearsley wrote:
> > > From latest run on android_nexus5_perf_bisect:
> > > 
> > > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> > > 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> > > cross-origin data at
> > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > and may not be loaded.
> > > documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
> > > taintsCanvas:0


+qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
android_nexus5_perf_bisect. Do you have idea?

video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
doc url: http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html

> > Thank you. Is there any LOG(ERROR) from webmediaplayer_android.cc ?
> 
> Nope. As mentioned earlier, I believe that logs from 
webmediaplayer_android.cc
> (or from anywhere in
> the Chrome renderer or browser processes) are not printed in the Telemetry
> output. But if someone can run the test locally, one can pass
> --enable--browser-logging to run_benchmark, and then the logs are output into
a
> temp file.

Ah, I partially missed your comment. Ok.

[Ken Russell (switch to Gerrit), 2016-03-02 19:44:30]

On 2016/03/02 18:50:55, dshwang wrote:
> On 2016/03/02 18:38:06, qyearsley wrote:
> > > > From latest run on android_nexus5_perf_bisect:
> > > > 
> > > > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> > > > 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> > > > cross-origin data at
> > > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > > and may not be loaded.
> > > >
documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
> > > > taintsCanvas:0
> 
> 
> +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
> android_nexus5_perf_bisect. Do you have idea?
> 
> video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
> doc url: http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html

What OS is this machine running? There was a problem with the Android media
player in earlier OS releases where Chrome couldn't guarantee that the played
video hadn't been redirected, so it marked all videos as cross-origin. I think
this was fixed in L but am not sure.

> > > Thank you. Is there any LOG(ERROR) from webmediaplayer_android.cc ?
> > 
> > Nope. As mentioned earlier, I believe that logs from 
> webmediaplayer_android.cc
> > (or from anywhere in
> > the Chrome renderer or browser processes) are not printed in the Telemetry
> > output. But if someone can run the test locally, one can pass
> > --enable--browser-logging to run_benchmark, and then the logs are output
into
> a
> > temp file.
> 
> Ah, I partially missed your comment. Ok.

[dshwang, 2016-03-02 21:22:37]

On 2016/03/02 19:44:30, Ken Russell wrote:
> On 2016/03/02 18:50:55, dshwang wrote:
> > On 2016/03/02 18:38:06, qyearsley wrote:
> > > > > From latest run on android_nexus5_perf_bisect:
> > > > > 
> > > > > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> > > > > 'texSubImage2D' on 'WebGLRenderingContext': The video element contains
> > > > > cross-origin data at
> > > > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > > > and may not be loaded.
> > > > >
> documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0 didPassCORSAccessCheck:0
> > > > > taintsCanvas:0
> > 
> > 
> > +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
> > android_nexus5_perf_bisect. Do you have idea?
> > 
> > video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
> > doc url: http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> 
> What OS is this machine running? There was a problem with the Android media
> player in earlier OS releases where Chrome couldn't guarantee that the played
> video hadn't been redirected, so it marked all videos as cross-origin. I think
> this was fixed in L but am not sure.

There is exact code which you say.
WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false if sdk revision is
less than 20
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...

SDK revision 20 is Android 4.1 (Jelly Bean) according to
http://android-developers.blogspot.fi/2012/06/android-sdk-tools-revision-20.html

qyearsley, could you check nexus 5 bot's android version? If it's lower than
Android 4.1, can you upgrade?

[qyearsley, 2016-03-02 22:00:44]

On 2016/03/02 21:22:37, dshwang wrote:
> On 2016/03/02 19:44:30, Ken Russell wrote:
> > On 2016/03/02 18:50:55, dshwang wrote:
> > > On 2016/03/02 18:38:06, qyearsley wrote:
> > > > > > From latest run on android_nexus5_perf_bisect:
> > > > > > 
> > > > > > FATAL: test fails due to GPU issue. SecurityError: Failed to execute
> > > > > > 'texSubImage2D' on 'WebGLRenderingContext': The video element
contains
> > > > > > cross-origin data at
> > > > > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > > > > and may not be loaded.
> > > > > >
> > documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0
didPassCORSAccessCheck:0
> > > > > > taintsCanvas:0
> > > 
> > > 
> > > +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false on
> > > android_nexus5_perf_bisect. Do you have idea?
> > > 
> > > video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
> > > doc url: http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > 
> > What OS is this machine running? There was a problem with the Android media
> > player in earlier OS releases where Chrome couldn't guarantee that the
played
> > video hadn't been redirected, so it marked all videos as cross-origin. I
think
> > this was fixed in L but am not sure.
> 
> There is exact code which you say.
> WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false if sdk revision
is
> less than 20
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...
> 
> SDK revision 20 is Android 4.1 (Jelly Bean) according to
>
http://android-developers.blogspot.fi/2012/06/android-sdk-tools-revision-20.html
> 
> qyearsley, could you check nexus 5 bot's android version? If it's lower than
> Android 4.1, can you upgrade?

Just checked the device android OS version, it is 4.4.2 (KOT49H).

[Ken Russell (switch to Gerrit), 2016-03-02 22:42:14]

On 2016/03/02 22:00:44, qyearsley wrote:
> On 2016/03/02 21:22:37, dshwang wrote:
> > On 2016/03/02 19:44:30, Ken Russell wrote:
> > > On 2016/03/02 18:50:55, dshwang wrote:
> > > > On 2016/03/02 18:38:06, qyearsley wrote:
> > > > > > > From latest run on android_nexus5_perf_bisect:
> > > > > > > 
> > > > > > > FATAL: test fails due to GPU issue. SecurityError: Failed to
execute
> > > > > > > 'texSubImage2D' on 'WebGLRenderingContext': The video element
> contains
> > > > > > > cross-origin data at
> > > > > > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > > > > > and may not be loaded.
> > > > > > >
> > > documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > > > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0
> didPassCORSAccessCheck:0
> > > > > > > taintsCanvas:0
> > > > 
> > > > 
> > > > +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false
on
> > > > android_nexus5_perf_bisect. Do you have idea?
> > > > 
> > > > video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
> > > > doc url: http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > 
> > > What OS is this machine running? There was a problem with the Android
media
> > > player in earlier OS releases where Chrome couldn't guarantee that the
> played
> > > video hadn't been redirected, so it marked all videos as cross-origin. I
> think
> > > this was fixed in L but am not sure.
> > 
> > There is exact code which you say.
> > WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false if sdk
revision
> is
> > less than 20
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...
> > 
> > SDK revision 20 is Android 4.1 (Jelly Bean) according to
> >
>
http://android-developers.blogspot.fi/2012/06/android-sdk-tools-revision-20.html
> > 
> > qyearsley, could you check nexus 5 bot's android version? If it's lower than
> > Android 4.1, can you upgrade?
> 
> Just checked the device android OS version, it is 4.4.2 (KOT49H).

Looking again, that check is highly suspect. It should be testing the run-time
Android version, not the SDK version against which the browser was built.

[dshwang, 2016-03-03 08:35:27]

On 2016/03/02 22:42:14, Ken Russell wrote:
> On 2016/03/02 22:00:44, qyearsley wrote:
> > On 2016/03/02 21:22:37, dshwang wrote:
> > > On 2016/03/02 19:44:30, Ken Russell wrote:
> > > > On 2016/03/02 18:50:55, dshwang wrote:
> > > > > On 2016/03/02 18:38:06, qyearsley wrote:
> > > > > > > > From latest run on android_nexus5_perf_bisect:
> > > > > > > > 
> > > > > > > > FATAL: test fails due to GPU issue. SecurityError: Failed to
> execute
> > > > > > > > 'texSubImage2D' on 'WebGLRenderingContext': The video element
> > contains
> > > > > > > > cross-origin data at
> > > > > > > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > > > > > > and may not be loaded.
> > > > > > > >
> > > >
documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > > > > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0
> > didPassCORSAccessCheck:0
> > > > > > > > taintsCanvas:0
> > > > > 
> > > > > 
> > > > > +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false
> on
> > > > > android_nexus5_perf_bisect. Do you have idea?
> > > > > 
> > > > > video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
> > > > > doc url:
http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > 
> > > > What OS is this machine running? There was a problem with the Android
> media
> > > > player in earlier OS releases where Chrome couldn't guarantee that the
> > played
> > > > video hadn't been redirected, so it marked all videos as cross-origin. I
> > think
> > > > this was fixed in L but am not sure.
> > > 
> > > There is exact code which you say.
> > > WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false if sdk
> revision
> > is
> > > less than 20
> > >
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...
> > > 
> > > SDK revision 20 is Android 4.1 (Jelly Bean) according to
> > >
> >
>
http://android-developers.blogspot.fi/2012/06/android-sdk-tools-revision-20.html
> > > 
> > > qyearsley, could you check nexus 5 bot's android version? If it's lower
than
> > > Android 4.1, can you upgrade?
> > 
> > Just checked the device android OS version, it is 4.4.2 (KOT49H).
> 
> Looking again, that check is highly suspect. It should be testing the run-time
> Android version, not the SDK version against which the browser was built.

kbr, there is --reduce-security-for-testing commend line.
I think we can reuse this command line to work around CORS check. What do you
think?
If you agree, I apply it in this CL.

[Ken Russell (switch to Gerrit), 2016-03-03 16:51:53]

On 2016/03/03 08:35:27, dshwang wrote:
> On 2016/03/02 22:42:14, Ken Russell wrote:
> > On 2016/03/02 22:00:44, qyearsley wrote:
> > > On 2016/03/02 21:22:37, dshwang wrote:
> > > > On 2016/03/02 19:44:30, Ken Russell wrote:
> > > > > On 2016/03/02 18:50:55, dshwang wrote:
> > > > > > On 2016/03/02 18:38:06, qyearsley wrote:
> > > > > > > > > From latest run on android_nexus5_perf_bisect:
> > > > > > > > > 
> > > > > > > > > FATAL: test fails due to GPU issue. SecurityError: Failed to
> > execute
> > > > > > > > > 'texSubImage2D' on 'WebGLRenderingContext': The video element
> > > contains
> > > > > > > > > cross-origin data at
> > > > > > > > http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4,
> > > > > > > > > and may not be loaded.
> > > > > > > > >
> > > > >
> documentUrl:http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > > > > > > hasWebMediaPlayer:1 hasSingleSecurityOrigin:0
> > > didPassCORSAccessCheck:0
> > > > > > > > > taintsCanvas:0
> > > > > > 
> > > > > > 
> > > > > > +qinmin, WebMediaPlayerAndroid::hasSingleSecurityOrigin() return
false
> > on
> > > > > > android_nexus5_perf_bisect. Do you have idea?
> > > > > > 
> > > > > > video url: http://127.0.0.1:44287/Canvas/resources/bear-1280x720.mp4
> > > > > > doc url:
> http://127.0.0.1:44287/Canvas/upload-video-to-sub-texture.html
> > > > > 
> > > > > What OS is this machine running? There was a problem with the Android
> > media
> > > > > player in earlier OS releases where Chrome couldn't guarantee that the
> > > played
> > > > > video hadn't been redirected, so it marked all videos as cross-origin.
I
> > > think
> > > > > this was fixed in L but am not sure.
> > > > 
> > > > There is exact code which you say.
> > > > WebMediaPlayerAndroid::hasSingleSecurityOrigin() return false if sdk
> > revision
> > > is
> > > > less than 20
> > > >
> > >
> >
>
https://code.google.com/p/chromium/codesearch#chromium/src/content/renderer/m...
> > > > 
> > > > SDK revision 20 is Android 4.1 (Jelly Bean) according to
> > > >
> > >
> >
>
http://android-developers.blogspot.fi/2012/06/android-sdk-tools-revision-20.html
> > > > 
> > > > qyearsley, could you check nexus 5 bot's android version? If it's lower
> than
> > > > Android 4.1, can you upgrade?
> > > 
> > > Just checked the device android OS version, it is 4.4.2 (KOT49H).
> > 
> > Looking again, that check is highly suspect. It should be testing the
run-time
> > Android version, not the SDK version against which the browser was built.
> 
> kbr, there is --reduce-security-for-testing commend line.
> I think we can reuse this command line to work around CORS check. What do you
> think?
> If you agree, I apply it in this CL.

Go ahead and try it to see if it works. However, I think we should really get to
the bottom of why the CORS check is failing. If the code's wrong then we need to
fix it!

[dshwang, 2016-03-04 16:27:43]

Patchset #10 (id:260001) has been deleted

[dshwang, 2016-03-04 16:30:51]

> > kbr, there is --reduce-security-for-testing commend line.
> > I think we can reuse this command line to work around CORS check. What do
you
> > think?
> > If you agree, I apply it in this CL.
> 
> Go ahead and try it to see if it works. However, I think we should really get
to
> the bottom of why the CORS check is failing. If the code's wrong then we need
to
> fix it!

I added --reduce-security-for-testing to blink_perf.py and makes the command
line skip single security origin check in android. 
In addition, file this issue to crbug.com/592017

In addition, I remain the change printing FATAL error because it's good.

kbr, could you review again?

https://codereview.chromium.org/1644353002/diff/280001/content/renderer/media...
File content/renderer/media/android/webmediaplayer_android.cc (right):

https://codereview.chromium.org/1644353002/diff/280001/content/renderer/media...
content/renderer/media/android/webmediaplayer_android.cc:759: return true;
skip single security origin check to work around nexus 5 issue.

https://codereview.chromium.org/1644353002/diff/280001/tools/perf/benchmarks/...
File tools/perf/benchmarks/blink_perf.py (right):

https://codereview.chromium.org/1644353002/diff/280001/tools/perf/benchmarks/...
tools/perf/benchmarks/blink_perf.py:93: '--reduce-security-for-testing'
add the command line

[dshwang, 2016-03-04 16:31:04]

The CQ bit was checked by dongseong.hwang@intel.com to run a CQ dry run

[dshwang, 2016-03-04 16:31:21]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707, 592017
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[commit-bot: I haz the power, 2016-03-04 16:31:26]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/280001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/280001

[Ken Russell (switch to Gerrit), 2016-03-04 18:21:43]

Very nice Dongseong. LGTM again.

[commit-bot: I haz the power, 2016-03-04 19:09:49]

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

[commit-bot: I haz the power, 2016-03-04 19:09:51]

Dry run: This issue passed the CQ dry run.

[aiolos (Not reviewing), 2016-03-04 19:24:53]

still lgtm

[dshwang, 2016-03-05 10:09:40]

dongseong.hwang@intel.com changed reviewers:
- qinmin@chromium.org

[dshwang, 2016-03-05 10:10:59]

dongseong.hwang@intel.com changed reviewers:
+ dalecurtis@chromium.org, qinmin@chromium.org

[dshwang, 2016-03-05 10:11:44]

qinmin, DaleCurtis, could you review webmediaplayer_android.cc ?

[DaleCurtis, 2016-03-07 18:36:58]

lgtm

[dshwang, 2016-03-07 19:54:48]

The CQ bit was checked by dongseong.hwang@intel.com

[dshwang, 2016-03-07 19:54:49]

The patchset sent to the CQ was uploaded after l-g-t-m from dtu@chromium.org,
qyearsley@chromium.org
Link to the patchset: https://codereview.chromium.org/1644353002/#ps280001
(title: "add --reduce-security-for-testing")

[commit-bot: I haz the power, 2016-03-07 19:55:09]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1644353002/280001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1644353002/280001

[dshwang, 2016-03-07 20:50:20]

On 2016/03/07 18:36:58, DaleCurtis wrote:
> lgtm

Thank you, Ken, qyearsley and all for persevering review and guide!

[commit-bot: I haz the power, 2016-03-08 00:04:49]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707, 592017
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707, 592017
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[commit-bot: I haz the power, 2016-03-08 00:04:51]

Committed patchset #10 (id:280001)

[commit-bot: I haz the power, 2016-03-08 00:07:01]

Description was changed from

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707, 592017
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Enable blink_perf.canvas on Android Perf bots

As the blocking issue was fixed, re-enable it.

BUG=496707, 592017
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:linux_perf_bisect;tryserver.chromium.perf:mac_10_10_perf_bisect;tryserver.chromium.perf:win_perf_bisect;tryserver.chromium.perf:android_nexus5_perf_bisect
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel

Committed: https://crrev.com/7d700edf761d46092b24d53ee228ce0df6e7b054
Cr-Commit-Position: refs/heads/master@{#379686}
==========

[commit-bot: I haz the power, 2016-03-08 00:07:05]

Patchset 10 (id:??) landed as
https://crrev.com/7d700edf761d46092b24d53ee228ce0df6e7b054
Cr-Commit-Position: refs/heads/master@{#379686}