All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=708666

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg7MnrlgsM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgnJ2g6goM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgnPOAqQkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg7P_bvgkM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg7I_IlggM


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

android-nexus5X
android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983134898605738128

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

Hi skym@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 : skym
  Commit : cd6b31d4b9613087e1a68c620af83b6fe1892bdf
  Date   : Wed Mar 29 06:47:00 2017
  Subject: [Sync] Stop accessing BrowserContextKeyedServiceFactory on non-UI thread.

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : memory.top_10_mobile_stress
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:java_heap:proportional_resident_size_avg/foreground/http_www_amazon_com_gp_aw_s_k_nexus
  Change       : 18.50% | 13105152.0 -> 15529984.0

Revision             Result                   N
chromium@460279      13105152 +- 1281851      6      good
chromium@460292      14278770 +- 5089121      9      good
chromium@460298      11919360 +- 1930852      6      good
chromium@460301      11060565 +- 1504059      6      good
chromium@460303      12561749 +- 7227994      6      good
chromium@460304      16387413 +- 884423       6      bad       <--
chromium@460329      15529984 +- 911794       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 memory.top_10_mobile_stress

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983134898605738128

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5249676711821312


| 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 Speed>Bisection.  Thank you!

I reverted my change on the 6th, see https://chromium.googlesource.com/chromium/src/+/a56dd49ba64abaf5300e5ccd8874447821e53da1

However, it doesn't look like the metrics have improved. Perhaps there was another regression that my changes hid? Unsure how to proceed, sending back to pmeenan.

Doesn't look like the blamed CL was responsible, and it's unlikely we could make progress on this so many months later.