All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=681590

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgoPCouwoM


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

android-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8990294572159565680

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : memory.top_10_mobile_stress
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_en_m_wikipedia_org_wiki_Science

Revision             Result                 N
chromium@443389      6412163 +- 425971      21      good
chromium@443443      6413733 +- 489310      21      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=http.en.m.wikipedia.org.wiki.Science memory.top_10_mobile_stress

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8990294572159565680

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5327210946232320


| 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!

Blocking on crbug.com/675639 for no memory.top_10_* repro

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8989548036508830944

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

Hi yangguo@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 : yangguo
  Commit : aa75904e3c2072f6b0d6ed92f75d1fb729d7cefa
  Date   : Thu Jan 12 14:18:45 2017
  Subject: [debugger] infrastructure for side-effect-free debug-evaluate.

Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : memory.top_10_mobile_stress
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_en_m_wikipedia_org_wiki_Science
  Change       : 0.39% | 6410526.28571 -> 6435340.85714

Revision                           Result                  N
chromium@443216                    6410526 +- 376708       14      good
chromium@443330                    6356556 +- 439946       9       good
chromium@443387                    6323574 +- 533234       14      good
chromium@443394                    6400444 +- 331047       14      good
chromium@443398                    6339006 +- 737143       21      good
chromium@443400                    6305031 +- 355285       6       good
chromium@443400,v8@2bca05685f      6347977 +- 529641       14      good
chromium@443400,v8@9884fb91e1      6379218 +- 536292       14      good
chromium@443400,v8@e00eae9e89      6246603 +- 3522465      21      good
chromium@443400,v8@aa75904e3c      6482176 +- 395802       14      bad       <--
chromium@443401                    6418601 +- 477479       14      bad
chromium@443415                    6413446 +- 615376       21      bad
chromium@443443                    6435341 +- 369379       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-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/8989548036508830944

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5908329450176512


| 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/8989496379715317696

4.6% is well within flakiness range. Restarting bisect.

=== BISECT JOB RESULTS ===
Perf regression found but unable to narrow commit range

Build failures prevented the bisect from narrowing the range further.


Bisect Details
  Configuration: android_nexus6_perf_bisect
  Benchmark    : memory.top_10_mobile_stress
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_en_m_wikipedia_org_wiki_Science
  Change       : 1.01% | 6398665.77778 -> 6463057.33333

Suspected Commit Range
  2 commits in range
  https://chromium.googlesource.com/chromium/src/+log/100c919f8fc51a06aee219a6898424f006cca92a..f5cb683f8e23e164b8ff7049d877fe266d4e8209


Revision             Result                  N
chromium@443388      6398666 +- 214805       9        good
chromium@443395      6405431 +- 279450       14       good
chromium@443399      6382469 +- 66359.0      14       good
chromium@443400      ---                     ---      build failure
chromium@443401      6446093 +- 211815       14       bad
chromium@443402      6433171 +- 13710.8      9        bad
chromium@443416      6447684 +- 93075.0      9        bad
chromium@443443      6463057 +- 167497       9        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=http.en.m.wikipedia.org.wiki.Science memory.top_10_mobile_stress

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8989496379715317696

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5278276337008640


| 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 bug has an owner, but was in a state that our triage picked up. Marking as Assigned.

Repeated bisect resulted in something entirely different, which confirms my suspicion that the first bisect is wrong. The graph seems to have recovered anyways.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8988398381744129184

Re #11: The difference between the bisects is:

# 3: Could not reproduce due to bug in bisect
# 6: Narrowed to your commit
# 9: Narrowed to the v8 roll with your commit in it, but the build of the v8 roll failed.

There are two graphs: the yandex one had some good runs, but didn't really recover, and the wikipedia one did not recover at all. All the bisects ran on the non-recovered wikipedia one. I ran the bisect on the yandex one.

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2969
Failure reason: the build has failed due to infrastructure failure.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8982620035238534688

This benchmark has been removed.