All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=779812

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=27740e1c3d6ad7bf721e03506711989a7d29561a72dca33dc3cf2b2fe425a865


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

android-nexus5
android-nexus5X
android-nexus6
android-one

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8964269731991985632

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

Hi npm@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 : Nicolas Pena
  Commit : dc09ff8f270573e210f6d388f2de134cacc3ed17
  Date   : Wed Oct 25 14:29:29 2017
  Subject: Reland: Add UMA histograms for EQT by task queue type

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : thread_times.tough_scrolling_cases
  Metric       : thread_raster_cpu_time_per_frame/text_constant_full_page_raster_60000_pixels_per_second
  Change       : 4.97% | 4.28391015927 -> 4.07084186576

Revision             Result                   N
chromium@511424      4.28391 +- 0.371198      9       good
chromium@511439      4.20372 +- 0.394771      14      good
chromium@511446      4.17736 +- 0.189254      14      good
chromium@511447      4.08034 +- 0.169614      6       bad       <--
chromium@511451      4.10357 +- 0.354413      14      bad
chromium@511453      4.06433 +- 0.416718      14      bad
chromium@511482      4.06864 +- 0.225308      9       bad
chromium@511540      4.07084 +- 0.258404      9       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=text.constant.full.page.raster.60000.pixels.per.second thread_times.tough_scrolling_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8964269731991985632


For feedback, file a bug with component Speed>Bisection

I wonder if this is due to increased UMA reporting or something else... We may need to think of another way to aggregate split EQT.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8964210030715217584

Kicking off another bisect just to be sure these results are correct.

This seems pretty surprising to me. Might be worth making a microbenchmark and profiling this code if it is legitimately the issue.

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

Suspected Commit
  Author : David Bokan
  Commit : dbab215523c8d77a7b54fcf3776357b079c36899
  Date   : Tue Oct 24 14:24:10 2017
  Subject: Fix telemetry gesture speeds

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : thread_times.tough_scrolling_cases
  Metric       : thread_total_all_cpu_time_per_frame/text_constant_full_page_raster_15000_pixels_per_second
  Change       : 7.49% | 26.3219950861 -> 28.2940512319

Revision                                 Result                   N
chromium@511393                          26.322 +- 1.69096        6      good
chromium@511396                          26.4831 +- 2.48988       6      good
chromium@511397                          26.3653 +- 1.57267       6      good
chromium@511397,catapult@f9691b256a      26.4048 +- 1.93158       6      good
chromium@511397,catapult@dbab215523      28.4617 +- 0.803745      6      bad       <--
chromium@511397,catapult@2f4cf37a9f      28.6626 +- 1.18022       6      bad
chromium@511398                          28.5473 +- 0.785091      6      bad
chromium@511402                          28.2943 +- 0.417701      6      bad
chromium@511410                          28.5918 +- 0.694347      6      bad
chromium@511426                          28.2941 +- 0.676518      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=text.constant.full.page.raster.15000.pixels.per.second thread_times.tough_scrolling_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8964210030715217584


For feedback, file a bug with component Speed>Bisection

Looks like it wasn't the EQT patch!