All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=677407

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg77PZtwsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg3-2LpQoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgv6LAoAoM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgv9DjvQoM


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

android-nexus5X
android-nexus9

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8991949163219169104

=== PERF REGRESSION ===


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

Hi mthiesse@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


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


===== SUSPECTED CL(s) =====
Subject : Fix leak in VRServiceImpl
Author  : mthiesse
Commit description:
  
This makes the implicit Strong Binding between the service and clients explicit, and fixes the service leaking when the binding is closed.

No functional changes here.

BUG= 651245 
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation
#TBR-ing a simple rename to match style.
TBR=brettw@chromium.org

Review-Url: https://codereview.chromium.org/2571323002
Cr-Commit-Position: refs/heads/master@{#439333}
Commit  : 7980a0e0ca7f5a6abda7e57ab28fbf8f845bfbf8
Date    : Sat Dec 17 07:38:21 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@439331  8986551  1664678  14  good
chromium@439332  8634587  2286251  14  good
chromium@439333  9227922  2341550  14  bad    <--
chromium@439335  9681920  172391   6   bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 677407

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.search.google system_health.memory_mobile
Test Metric: memory:chrome:all_processes:reported_by_os:system_memory:java_heap:proportional_resident_size_avg/load_search/load_search_google
Relative Change: 2.17%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/1024
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991949163219169104


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5901972660551680

| 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!

The graphs look to me like this is just noise and not actually a regression. I also don't think my CL could have caused a memory increase (it should actually lower it over time as it fixes a leak).

I can't actually get tools\perf\run_benchmark to run locally to verify. I get error messages like "Cannot get standard output on Android" and "Tracing is not started".

If somebody insists this is actually a regression I'll spend more time looking into it, but for now I think there's no regression here.

ccing dcheng who suggested the change, in case I missed something.