All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=774102

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=2f12036b831d212bc6077ea8d276846618b998da2a2e52960118b9a358f445c8


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

android-nexus7v2

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965932116139995792

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

Hi altimin@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 : Alexander Timin
  Commit : aee6977a4f0324b333cc16d58f6fa9d95c097e8e
  Date   : Tue Oct 10 14:09:16 2017
  Subject: [scheduler] Record time each time an immediate task is posted.

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : thread_times.tough_compositor_cases
  Metric       : tasks_per_frame_raster/tasks_per_frame_raster
  Change       : 1.79% | 4.75085652738 -> 4.83589100476

Revision             Result                    N
chromium@507650      4.75086 +- 0.0234437      6      good
chromium@507656      4.746 +- 0.0399125        6      good
chromium@507658      4.72871 +- 0.0262381      6      good
chromium@507659      4.81985 +- 0.0332203      6      bad       <--
chromium@507661      4.81458 +- 0.0358077      6      bad
chromium@507671      4.81808 +- 0.0331154      6      bad
chromium@507691      4.84238 +- 0.0453296      6      bad
chromium@507731      4.83589 +- 0.0511466      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 thread_times.tough_compositor_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965932116139995792


For feedback, file a bug with component Speed>Bisection

altimin: looks like this regression is real, but very small and only on a few synthetic test cases on a single device. Closing, but feel free to reopen if you're still working on it.