All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=780232

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=0a64555c1d24f1d951325059b99c0ddea8df9c9ebbbea8774e21d933ebf07c8d


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

android-nexus5
android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8964184406853032128

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

Hi nverne@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 : Nicholas Verne
  Commit : 5420564de0e2b7da728e7a9402e04b6cdfb9fc0a
  Date   : Fri Oct 27 07:37:54 2017
  Subject: Mojoify control messages for child processes.

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_chrome:malloc:effective_size_avg/browse_chrome/browse_chrome_newtab
  Change       : 11.87% | 49820263.6667 -> 55735967.6667

Revision             Result                   N
chromium@512067      49820264 +- 2038835      6      good
chromium@512095      49569746 +- 1062412      6      good
chromium@512109      49223105 +- 1414391      6      good
chromium@512113      48727691 +- 937032       6      good
chromium@512114      49084523 +- 1185524      6      good
chromium@512115      54948856 +- 281229       6      bad       <--
chromium@512116      55144980 +- 1803016      6      bad
chromium@512122      55735968 +- 1149256      6      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

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=browse.chrome.newtab system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8964184406853032128


For feedback, file a bug with component Speed>Bisection

 Issue 780230  has been merged into this issue.

 Issue 780231  has been merged into this issue.

nverne: ping on this? The bisect identified a 5MiB regression on this bug and a 17MiB regression on  bug 780231 , which are massive amounts for Android (more than we get in gains from multi-quarter memory improvement projects).

cc-ing the CL reviewers and test owners as well.

It looks like the metrics recovered some time around?
https://chromium.googlesource.com/chromium/src/+log/7020e113f699fcc251061589e272257931c0fafe%5E..a86f030e69d09286b36ec9e52da96334f5aa6c95?pretty=oneline&n=1000

Yes. That would be
https://chromium.googlesource.com/chromium/src/+/46bed17ef6f36d5ef97a9e4700cdc035b1cddc16
which fixed leaking child processes.

And to clarify, the leaking child processes were in fact introduced by the
culprit CL here. So I think we can say it's fixed.

Awesome, sounds good! Thanks for checking!