All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=760568

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=c0516861d5fe42acce680235d17c07b5741e48f9e779476bca6ef904a8e82731


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8969831186076218672

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : smoothness.sync_scroll.key_mobile_sites_smooth
  Metric       : mean_input_event_latency/https___www.google.co.uk_search?hl_en_q_barack+obama_cad_h


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=https...www.google.co.uk.search.hl.en.q.barack.obama.cad.h smoothness.sync_scroll.key_mobile_sites_smooth

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8969831186076218672


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965375082707786800

It looks like the culprit was reverted (or revert/reland/revert), but this seems to have landed before M62 branch cut and been reverted after. Bisecting in case the fix is something we should consider for a merge.

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

Hi timloh@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 : Timothy Loh
  Commit : 2db3fa641f2b6a05ee894f158a93a419caf3e896
  Date   : Tue Aug 29 03:09:33 2017
  Subject: Prepare tests for enabling modal permission prompts and remove gesture experiment

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : smoothness.sync_scroll.key_mobile_sites_smooth
  Metric       : mean_input_event_latency/https___www.google.co.uk_search?hl_en_q_barack+obama_cad_h
  Change       : 237.20% | 28.7358333333 -> 96.8981666667

Revision             Result                  N
chromium@497705      28.7358 +- 11.5433      6      good
chromium@497861      26.2102 +- 3.10071      6      good
chromium@497939      27.3882 +- 8.63249      6      good
chromium@497978      28.5095 +- 8.82504      6      good
chromium@497997      29.0372 +- 9.54519      6      good
chromium@498002      31.7173 +- 18.5178      6      good
chromium@498003      30.2845 +- 12.4666      6      good
chromium@498004      109.009 +- 77.241       6      bad       <--
chromium@498005      122.909 +- 37.3136      6      bad
chromium@498007      122.984 +- 99.1976      6      bad
chromium@498016      96.8982 +- 51.1218      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=https...www.google.co.uk.search.hl.en.q.barack.obama.cad.h smoothness.sync_scroll.key_mobile_sites_smooth

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965375082707786800


For feedback, file a bug with component Speed>Bisection

The culprit CL is a test change, definitely doesn't need a merge.

Some context at  bug 762335 .