All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=702002

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgtKOkswkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9NHepQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgtN_k6woM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9JiWrwkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9LarpQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9Nbk8wgM


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

android-nexus5
android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8985015141676182224

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : board_temperature/video.html?src_tulip2.m4a_type_audio

Revision             Result                  N
chromium@456678      33166.7 +- 2422.12      21      good
chromium@456713      33361.9 +- 1212.24      21      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

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8985015141676182224

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5108367908929536


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984949228095706528

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984949218637401984

First bisect was for board temp - no findings. Bisects in comment 4/5 are for vm_shared_dirty_delta. Bisects in comments 7 and 8 will focus on working_set_delta_size

Looking closer, the board temperature and working_set_delta seem to be noise - ref graphs see similar spikes. Ignored those and will focus just on vm_shared_dirty_delta going forward.

Also, this regressions range overlaps other regressed metrics for windows bots
 Issue 702337  - vm_working_set
 Issue 702332  - time-top-play 
 Issue 702268  - process_delta

Unclear if issues are related.  

Graphs have recovered.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984378643898913248

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984378629933883856

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

Hi steimel@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 : steimel
  Commit : 5b069f0bfe05879dc2dfbc32d9b4b171c622d5c3
  Date   : Tue Mar 14 16:25:35 2017
  Subject: Hide overlay play button if it can't be shown without clipping.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : vm_shared_dirty_delta/crowd1080_vp9.webm_gpu
  Change       : 3.56% | 14923.6666667 -> 15454.6666667

Revision             Result                  N
chromium@456684      14923.7 +- 32.5474      6      good
chromium@456712      14916.3 +- 43.7645      6      good
chromium@456719      14921.3 +- 39.3616      6      good
chromium@456723      14914.7 +- 50.964       6      good
chromium@456724      15450.0 +- 23.8328      6      bad       <--
chromium@456725      15457.3 +- 33.4265      6      bad
chromium@456726      15451.3 +- 58.901       6      bad
chromium@456740      15454.7 +- 28.2371      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 media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984378643898913248

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4874103212736512


| 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!

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

Suspected Commit
  Author : steimel
  Commit : 5b069f0bfe05879dc2dfbc32d9b4b171c622d5c3
  Date   : Tue Mar 14 16:25:35 2017
  Subject: Hide overlay play button if it can't be shown without clipping.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : vm_shared_dirty_delta/crowd1080_vp9.webm_total
  Change       : 2.70% | 19523.3333333 -> 20051.3333333

Revision             Result                  N
chromium@456684      19523.3 +- 43.2358      6      good
chromium@456712      19514.9 +- 52.1427      7      good
chromium@456719      19496.3 +- 43.5813      6      good
chromium@456723      19523.6 +- 38.9256      5      good
chromium@456724      20061.7 +- 61.6063      6      bad       <--
chromium@456725      20060.7 +- 42.395       6      bad
chromium@456726      20048.7 +- 63.9792      6      bad
chromium@456740      20051.3 +- 72.148       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 media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984378629933883856

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4923065839910912


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8982651975586946144