All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=702848

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg9KbtsgsM


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

android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984827129851650560

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

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : cpu_utilization/video.html?src_crowd1080.ogv_canvas_true

Revision             Result                   N
chromium@457387      2.70549 +- 0.803214      21      good
chromium@457433      2.7005 +- 0.543891       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.crowd1080.ogv.canvas.true media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984827129851650560

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6706417462935552


| 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/8984581486189053248

Both values in comment 3 bisect are in regressed range. Doing new bisect with earlier start.

Also, I notice this regression is really a return to previous averages after an improvement dip around point 456475. Will bisect the improvement as well. If neither improved/regressed points are confirmed by bisect, will resolve as no-repro.

Improvement is being bisected here:
https://chromeperf.appspot.com/buildbucket_job_status/8984485080010177760

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

Suspected Commit
  Author : chaopeng
  Commit : c7a846538699afcb05ba89a8dd8abbf5bbc418ab
  Date   : Thu Mar 16 03:16:08 2017
  Subject: Revert of Show Overlay Scrollbar when GestureScrollUpdate (patchset #10 id:260001 of https://codereview.chromium.org/2611013002/ )

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : cpu_utilization/video.html?src_crowd1080.ogv_canvas_true
  Change       : 46.51% | 1.80083812237 -> 2.63835117007

Revision             Result                   N
chromium@457200      1.80084 +- 0.27958       6      good
chromium@457317      1.75701 +- 0.267794      6      good
chromium@457332      1.92391 +- 0.222537      6      good
chromium@457333      1.83046 +- 0.411948      6      good
chromium@457334      2.67392 +- 0.135384      6      bad       <--
chromium@457336      2.70707 +- 0.320046      6      bad
chromium@457339      2.78039 +- 0.47883       6      bad
chromium@457346      2.69115 +- 0.407118      6      bad
chromium@457375      2.66798 +- 0.389241      6      bad
chromium@457433      2.63835 +- 0.471063      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.crowd1080.ogv.canvas.true media.android.tough_video_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984581486189053248

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5114979071557632


| 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!