All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=798376

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=cf24944e93be54c1bcd9e82d3d3ee48aa0f564868278f8ebcb86651eb752a825


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

android-nexus5

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8958512773486294336

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

Hi chrishtr@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 : Chris Harrelson
  Commit : d271bad92f0eadddc4fd92d0a99ad5c7ae77b5dd
  Date   : Thu Dec 28 21:59:32 2017
  Subject: [PE] Fix code that dirties compositing inputs when overflow changes.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : thread_times.tough_compositor_cases
  Metric       : thread_renderer_main_cpu_time_per_frame/thread_renderer_main_cpu_time_per_frame
  Change       : 4.98% | 2.97373984039 -> 3.12189688862

Revision             Result                    N
chromium@526328      2.97374 +- 0.0135405      6      good
chromium@526329      2.98408 +- 0.0241082      6      good
chromium@526330      3.14269 +- 0.0263614      6      bad       <--
chromium@526331      3.13786 +- 0.0154065      6      bad
chromium@526333      3.14233 +- 0.023999       6      bad
chromium@526337      3.15035 +- 0.0266765      6      bad
chromium@526345      3.15801 +- 0.0240946      6      bad
chromium@526361      3.1219 +- 0.0298724       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/8958512773486294336


For feedback, file a bug with component Speed>Bisection

This regression doesn't seem to make much sense, unless there is hit
testing triggered by this indirectly.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8958297739332358336

Re-bisecting.

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

Suspected Commit
  Author : Chris Harrelson
  Commit : d271bad92f0eadddc4fd92d0a99ad5c7ae77b5dd
  Date   : Thu Dec 28 21:59:32 2017
  Subject: [PE] Fix code that dirties compositing inputs when overflow changes.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : thread_times.tough_compositor_cases
  Metric       : thread_renderer_main_cpu_time_per_frame/thread_renderer_main_cpu_time_per_frame
  Change       : 5.00% | 2.94865856331 -> 3.09616529491

Revision             Result                     N
chromium@526328      2.94866 +- 0.0345801       6      good
chromium@526329      2.96172 +- 0.0255919       6      good
chromium@526330      3.1003 +- 0.0227479        6      bad       <--
chromium@526331      3.11439 +- 0.0409619       6      bad
chromium@526333      3.10933 +- 0.0281633       6      bad
chromium@526337      3.12191 +- 0.031453        6      bad
chromium@526345      3.12387 +- 0.0188809       6      bad
chromium@526361      3.09617 +- 0.00972287      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/8958297739332358336


For feedback, file a bug with component Speed>Bisection

The regression has recovered. Not sure why. Bisecting.