All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=626516

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg0tve5gsM


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

android-nexus5X

Probably https://codereview.chromium.org/2123653005 ?
Cc-ing drott@

It seems that Pinterest is affected much.

https://chromeperf.appspot.com/report?sid=fe3343a6097fd5a14df1a5f437059fa3e48660ccb24bf575ca9e903611ce28fb
 

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


===== SUSPECTED CL(s) =====
Subject : Activate complex text path on Android
Author  : drott
Commit description:
  
In Chromium commit 10749067720a341 catapult was rolled, switching to
default teardowns between pageset repeats, compare [1]. Activating
complex text on Android again after this change in order to see how
measurements are affected by this. If no memory regression shows up in
the memory health measurements for memory pagesets this would indicate
that we may have a memory-leakage over time problem in webview
single-process mode but no issues in the usual tab process lifecycle on
Android. Activating and TBR'ing to collect bot results, potentially
reverting later if needed for further investigation.

[1] https://github.com/catapult-project/catapult/issues/2294

BUG=577306
TBR=eae

Review-Url: https://codereview.chromium.org/2123653005
Cr-Commit-Position: refs/heads/master@{#403884}
Commit  : 5009e3effbb04c74140d262fb71bc51cbf38913a
Date    : Wed Jul 06 11:39:32 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@403880  1485.6   8.76356  5  good
chromium@403883  1484.8   16.0997  5  good
chromium@403884  92345.6  118.802  5  bad    <--
chromium@403885  92310.2  254.513  5  bad
chromium@403889  92426.4  99.6534  5  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 626516

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests memory.blink_memory_mobile
Test Metric: memory:chrome:renderer_processes:reported_by_chrome:font_caches:effective_size_avg/memory:chrome:renderer_processes:reported_by_chrome:font_caches:effective_size_avg
Relative Change: 6121.49%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/308
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9007748833708237360


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5880131382411264

| 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!

This effect can be likely marked as WontFix since the value is only used in the complex font path and we don't have equivalent memory reporting in the simple path.