All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=690224

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg6LK7sgoM


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

android-webview-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8988182116873257072

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

Hi jgruber@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 : jgruber
  Commit : cb19ecd610ee0e7f8bc661616cdb9ca4e9ea804b
  Date   : Thu Feb 02 11:31:01 2017
  Subject: [string] Migrate String.prototype.{split,replace} to TF

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:malloc:effective_size_avg/load_search/load_search_google
  Change       : 1.03% | 22611218.2857 -> 22843361.7143

Revision                           Result                  N
chromium@448217                    22611218 +- 786951      14      good
chromium@448222                    22643686 +- 587157      14      good
chromium@448224                    22580145 +- 821495      14      good
chromium@448225                    22643238 +- 294606      9       good
chromium@448225,v8@cb19ecd610      22924464 +- 262408      6       bad       <--
chromium@448225,v8@13de00c0be      22992916 +- 234576      6       bad
chromium@448225,v8@c85ff82822      22929975 +- 660088      14      bad
chromium@448225,v8@8ff508735e      22924056 +- 406270      9       bad
chromium@448225,v8@d0edd08eb5      22984508 +- 199828      6       bad
chromium@448225,v8@dbda66ec4f      22992351 +- 584497      9       bad
chromium@448226                    22937713 +- 589868      9       bad
chromium@448235                    22843362 +- 441507      14      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.search.google system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8988182116873257072

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5276409418743808


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8988152601720504352

I don't see what that CL has to do with malloc'ed memory, restarting the bisect.

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : jgruber
  Commit : cb19ecd610ee0e7f8bc661616cdb9ca4e9ea804b
  Date   : Thu Feb 02 11:31:01 2017
  Subject: [string] Migrate String.prototype.{split,replace} to TF

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:browser_process:reported_by_chrome:malloc:effective_size_avg/load_search/load_search_google
  Change       : 1.66% | 22545372.0 -> 22867606.6667

Revision                           Result                  N
chromium@448217                    22545372 +- 558587      14      good
chromium@448222                    22619398 +- 819250      14      good
chromium@448224                    22681460 +- 936076      14      good
chromium@448225                    22665833 +- 503050      14      good
chromium@448225,v8@cb19ecd610      23069335 +- 382624      6       bad       <--
chromium@448225,v8@13de00c0be      23055243 +- 444896      6       bad
chromium@448225,v8@c85ff82822      22918540 +- 832726      14      bad
chromium@448225,v8@8ff508735e      23005594 +- 264109      9       bad
chromium@448225,v8@d0edd08eb5      22940955 +- 353770      6       bad
chromium@448225,v8@dbda66ec4f      22880572 +- 520151      14      bad
chromium@448226                    22907341 +- 608007      14      bad
chromium@448235                    22867607 +- 240048      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.search.google system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8988152601720504352

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5812836606935040


| 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!

Had another look at this and I still don't see a relation to the blamed CL. Graphs at [0] show no jumps in V8 malloc'ed memory, and no regressions at the same point on other bots.

[0] https://chromeperf.appspot.com/report?sid=d70d32f49b2dfd57c279c184a03f384e1b94a52eb7c380ef5cf280bc364ffc59