All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=645379

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgieTitwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8aabtAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8Zb6vgoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8dvprQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgiaLRsAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgicb-uAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgib7XuwoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgid2OuAkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg8aWptQsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgiZLfsQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgiYGZ7QgM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg0amyyQgM


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

android-galaxy-s5
android-nexus5
android-nexus5X
android-nexus7v2
android-nexus9
android-one

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9002017431171322704

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

Hi alexclarke@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Remove after wakeup logic and replace PumpTask with Fences
Author  : alexclarke
Commit description:
  
Task pumping is a neat concept for throttling but it adds a lot of
complexity to the task queues.  For example it prevents the WorkQueues
from being read only (which they could otherwise be).

Fences provide similar functionality but don't preclude
various queue optimizations.  They are also a more familiar
concept which should make the code easier to comprehend.

The after wake up logic isn't used and it also adds a lot of
complexity. Lets get rid of it :)

BUG= 638542 

Review-Url: https://codereview.chromium.org/2276353002
Cr-Commit-Position: refs/heads/master@{#416558}
Commit  : c2db8aa78e1130a08bd47f2aa611f322afaf8dc5
Date    : Mon Sep 05 18:06:10 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev     N  Good?
chromium@416548  2.1138   0.00612011  5  good
chromium@416554  2.11155  0.00947613  5  good
chromium@416557  2.11427  0.00970767  5  good
chromium@416558  3.11592  0.0149972   5  bad    <--
chromium@416559  3.10227  0.0103867   5  bad

Bisect job ran on: android_one_perf_bisect
Bug ID: 645379

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_idle_power_cases
Test Metric: tasks_per_second_total_all/set-timeout.html (Long Idle)
Relative Change: 46.76%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_one_perf_bisect/builds/1609
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9002017431171322704


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5881549776486400

| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Tests>AutoBisect.  Thank you!

I suspect this benchmark is broken.  If the number of tasks per second had gone up I would expect to see that in the traces, but I don't.

Consider Trace A (before the regression) MessageLoop:RunTask is called 205 times during the Interaction.IdleWaiting. Trace A (after the regression) MessageLoop:RunTask is called 202 times during the Interaction.IdleWaiting.

That sounds like it's gone down slightly (likely noise rather than a real change).


Trace A: https://00e9e64bac42627a5f059c46c4c76ed5649364250ca017061b-apidata.googleusercontent.com/download/storage/v1_internal/b/chrome-telemetry-output/o/trace-file-id_5-2016-09-06_12-29-22-63227.html?qk=AD5uMEuNYkMz3qNPycFZ9bn077Ed4MC69ZOfyraPBQ-IY7FbcWt_mktWgoFa0eexReJMpvszaHB2Q4i5Ou2yfJYR85CyG8792vOKEpl4SAdJMx-YV3MSIiIiVkMbAL1wYWt5x6A4QjAwf9pDQx-myORcGIiQbwizXIrvOQK5MUUaGNMz8_o2f7WP88I1FlmE8fF9WyvlNEo-PdwqrkZyaC_zHHsPQhik0MiCylOnMaI9xKbW1m0X2QLTNwOlKuxL_ZXRx5CG3vmIHrI9uIUvBoJhCzjmnJu7_1J9F_b9PPRzwhx31DmOXNBUySET5H3x5wueVcUJKhAPYAyaB7A2Lltl2a3dNBhZJSDSE2YqCzdF8oNa83Hcp_5Yjp1hkFLZAPs-x_YAD7l1V7RvJ4vugi93BLj6xgS5BtQCXApQOCt0aMzmWFZmVE-IwZLpd2_EZDQSeGgiVYOXN94_BIXyGfikBEow91cuqG8JWWt3V5SyLrDTPFd6pybHGJL9vuL6oh8etw257-Zby5Ae99vmNP50IG7IIn0LHfWc48OSr30NtUoYENxnnsCw7k0iaJUPWFk0YMabm5pOBg_WdsYTjxSJbyITTgB6EBnqL_y0XX-69eJgnoOrtN6xpU9G39FIqPSHsJl4xm1wJpaOCh23ivAuB8mZdxxgDTvzzK2VtOrCH2jVeZXj9mZB6hbYC5AuL91CLJm9CpYK9AbS1vaFoivNtnPB4r8sai-zRC04ryy5fOdSqnUgb63bJUUz1cjsJ91IPUfa_xs_ajTcuwRnUlbe7vTIuBX4fRLDNZOSqWY_uxQz1dswuPb1z7WdNLl2JxhTcd9SbSGU8xA1rthRTM3t2XQwdawNEyUsZCSAUECu-fuqBGV6yCs


Trace B: https://00e9e64bac7aa16157795d339ac49405c40d3b93ec0800e9ae-apidata.googleusercontent.com/download/storage/v1_internal/b/chrome-telemetry-output/o/trace-file-id_5-2016-09-07_08-33-51-35533.html?qk=AD5uMEsOoTMGVuHB2RvvQYujE-tbhRTSlmQK2QAGIVilCuEA64PnDSqZPgYs6pca0YaWFydfjLY93N8N1VgiKvX-8TL4BjwxoYeQ2rqqz9oI2zxXeqArzDIV1slb5ZWcWUvHAV9yiCVB-ZP6MQ-zqIUiAc84LdIC4HeNDIBxWyz2w8uAfjXEkKkh0bPB34J8aO1EVClSrXQhM6qFy-tPeYFYktky8vciFjxUp17CsR5gS3hmW9jkKsh3gkGyCSqB5hw38XGBf_6wj71cc8S11danb71OIe_X2mpM2F6eIHjMLbf0bnGK576zMi6xbPbYcim_1XUGlCizX8YuJn-dalA-ZChEjLYOsOFKBHMgFpuJNxyAPW3E-nyW3b37wrrdi_FePnTnRnYHJvF48YVPwju90KEgcSE80HIDjBcrMzoD7edNmRw32LUN8XNS9ipkCJ6CxeOdwBbjHtZV-2jaEBKHWS0aO5vKMedIFPd6EJy5rzgoe08kIVu4W2zvJuNiMpYWTrXHoo0lDmr1ERg0bVyIWAElE9hfyQBjyfaYQoCh334iruSsuest7o8v-h6pOK6GEZcosSngJdkMjr03DmI6Xja0FQk9kDILCKL0j-zE-aU9CqXecl_mcD4gbzCuWd24r-OXOnxaEMjLTdRyJAVjYzcxpRXKeniTov4ue72Clge1diQY3Obd0GZ70J-h-eFJ6hCxn7lnKUT6oFp4-sPFjtbsvgGpsMFA8_Eg0-dNZ4MeExVh_92DjhxSPc0G6dBEJjfSu00GwYq1ngHL5vKgQIU34U4VcvGAcFNWh1UVziHzRhyuEXp3jzeqBMzaIsapL3LwskeCWp8K6vyeFLrnOLHT4cIAnISPeuNse7vtmXMCX3AejJQ