All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=759557

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=a1db0f4eaebb6eae52fb3ebd1a20dc50b4fbdccf33476c54164b6e65940e84bf


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

android-webview-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8970010632907121312

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

Hi asvitkine@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 : Alexei Svitkine (very slow)
  Commit : 43aa08eb002ccb00174fefd9be693498a0d58a5c
  Date   : Fri Aug 25 23:31:09 2017
  Subject: Revert "Validate histograms before and after each task on Android."

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : thread_times.key_silk_cases
  Metric       : thread_raster_cpu_time_per_frame/masonry.html
  Change       : 16.64% | 2.05109707138 -> 2.3924384177

Revision             Result                    N
chromium@497553      2.0511 +- 0.0953789       6      good
chromium@497580      2.08546 +- 0.112792       6      good
chromium@497587      2.07886 +- 0.0783059      6      good
chromium@497591      2.10909 +- 0.116015       6      good
chromium@497593      2.08208 +- 0.0568977      6      good
chromium@497594      2.37131 +- 0.145968       6      bad       <--
chromium@497607      2.37176 +- 0.0748961      6      bad
chromium@497661      2.39244 +- 0.0810367      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=masonry.html thread_times.key_silk_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8970010632907121312


For feedback, file a bug with component Speed>Bisection

Hey kraynov@ - the CL reverts instrumentation that should slow things down. So it's surprising it caused a regression - it should cause an improvement.

kraynov@ - could you check whether the test expectations are backwards here and that this is not actually an improvement?

Issue 759570 has been merged into this issue.

+vmiura: any ideas on #4?