All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=727496

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgxtSsqgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-u_oowkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-o7trQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgur_inAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDghsCvtQsM


Bot(s) for this bug's original alert(s):

android-nexus5
android-nexus7v2

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8978195800352030752

=== Auto-CCing suspected CL author klausw@chromium.org ===

Hi klausw@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : klausw
  Commit : 89f6d708d1be7d2c1cebaa74b11908740130f617
  Date   : Thu May 25 18:57:24 2017
  Subject: WebVR: use optimized render path with deferred wait

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : media.android.tough_video_cases_tbmv2
  Metric       : cpu_time_percentage_avg/video.html?src_tulip2.mp3_type_audio_seek
  Change       : 8.66% | 0.645306639617 -> 0.701161089825

Revision             Result                     N
chromium@474670      0.645307 +- 0.195618       14      good
chromium@474711      0.642015 +- 0.0656938      6       good
chromium@474723      0.684758 +- 0.177993       14      good
chromium@474729      0.681961 +- 0.0640444      9       good
chromium@474731      0.679359 +- 0.0991512      9       good
chromium@474732      0.661363 +- 0.0798478      6       good
chromium@474733      0.728245 +- 0.11214        14      bad       <--
chromium@474753      0.707389 +- 0.120057       14      bad
chromium@474835      0.701161 +- 0.116028       14      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=video.html.src.tulip2.mp3.type.audio.seek media.android.tough_video_cases_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978195800352030752

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5850331868561408


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!

I strongly doubt that this is related to my change - the blamed CL only edits the single file third_party/WebKit/Source/modules/vr/VRDisplay.cpp which is only used for WebVR, and this code should not be active at all for standard 2D video playback.

Since the graphs are fairly noisy, is it possible that the actual culprit is one of the succeeding CLs?

Reassigning to toyoshim@ for next steps.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8978088755117389328

Kicked another bisect with a wider revision range, on another device, against another test.

=== Auto-CCing suspected CL author fsamuel@chromium.org ===

Hi fsamuel@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : fsamuel
  Commit : d71466ad118aca05aa347c9766f80a5f77528588
  Date   : Thu May 25 20:35:22 2017
  Subject: cc::SurfaceDependencyTracker should not crash when a Display goes away

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : media.android.tough_video_cases_tbmv2
  Metric       : cpu_time_percentage_avg/video.html?src_tulip2.m4a_type_audio
  Change       : 11.33% | 0.219632635077 -> 0.244513086249

Revision             Result                      N
chromium@474333      0.219633 +- 0.0130212       6      good
chromium@474710      0.223762 +- 0.00460524      6      good
chromium@474757      0.235279 +- 0.0047822       6      good
chromium@474781      0.23526 +- 0.00688859       6      good
chromium@474782      0.237923 +- 0.00601247      6      good
chromium@474783      0.24752 +- 0.00902954       6      bad       <--
chromium@474784      0.242584 +- 0.00824214      6      bad
chromium@474787      0.246786 +- 0.0080162       6      bad
chromium@474793      0.244224 +- 0.00703015      6      bad
chromium@474804      0.247161 +- 0.0100316       6      bad
chromium@474898      0.245746 +- 0.00332187      6      bad
chromium@475086      0.244513 +- 0.00658637      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=video.html.src.tulip2.m4a.type.audio media.android.tough_video_cases_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978088755117389328

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6366766520860672


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!