All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=681541

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgwKv8rgsM


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

android-one

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8990306387325281456

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8990305261781336384

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

Hi mvstanton@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 : mvstanton
  Commit : b8294aaa978e972978b8d82cedf63befad5d7af1
  Date   : Thu Jan 12 11:29:09 2017
  Subject: [TypeFeedbackVector] Root literal arrays in function literals slots

Bisect Details
  Configuration: android_one_perf_bisect
  Benchmark    : v8.browsing_mobile
  Metric       : v8-gc-scavenger_sum/browse_social/browse_social_facebook
  Change       : 16.95% | 362.256 -> 423.670833333

Revision                           Result                  N
chromium@443295                    362.256 +- 23.8185      6      good
chromium@443369                    365.356 +- 26.0955      6      good
chromium@443388                    376.744 +- 24.1847      6      good
chromium@443398                    370.285 +- 18.0092      6      good
chromium@443400                    373.855 +- 32.0527      6      good
chromium@443400,v8@2bca05685f      378.218 +- 25.3568      6      good
chromium@443400,v8@b8294aaa97      415.289 +- 40.0038      6      bad       <--
chromium@443400,v8@81c62e070b      423.13 +- 35.6014       6      bad
chromium@443400,v8@9884fb91e1      439.275 +- 33.1123      6      bad
chromium@443400,v8@aa75904e3c      417.158 +- 22.8104      6      bad
chromium@443401                    428.091 +- 54.3194      6      bad
chromium@443403                    416.211 +- 43.5471      6      bad
chromium@443407                    431.421 +- 41.1539      6      bad
chromium@443443                    423.671 +- 26.7592      6      bad

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=browse.social.facebook v8.browsing_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8990306387325281456

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5271205109563392


| 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!

=== 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: mac_retina_perf_bisect
  Benchmark    : v8.browsing_desktop
  Metric       : v8-gc-scavenger_avg/browse_news/browse_news_nytimes
  Change       : 11.06% | 8.10914124113 -> 9.00641516936

Suspected Commit Range
  4 commits in range
  Mismatching LKGR/FKBR depots, unable to provide handy url.
  good_revision: chromium@05574f4462f910edf41883479fcffabc21e2198f
  bad_revision : v8@cfc2e5e180f416e24fcbf532f32ded47d38a49ff


Revision                           Result                   N
chromium@443256                    8.10914 +- 0.709859      6        good
chromium@443356                    8.44739 +- 0.717412      9        good
chromium@443382                    8.43196 +- 0.330028      6        good
chromium@443395                    8.3881 +- 0.495185       6        good
chromium@443398                    8.29184 +- 0.709932      6        good
chromium@443400                    8.40652 +- 0.70132       14       good
chromium@443400,v8@51c5d91393      ---                      ---      build failure
chromium@443400,v8@1adc70b0f9      ---                      ---      build failure
chromium@443400,v8@3188780410      ---                      ---      build failure
chromium@443400,v8@cfc2e5e180      8.96404 +- 0.27959       4        bad
chromium@443400,v8@b8294aaa97      8.88776 +- 0.29689       4        bad
chromium@443400,v8@aa75904e3c      9.28173 +- 0.471468      4        bad
chromium@443401                    9.03986 +- 0.830187      9        bad
chromium@443407                    9.08583 +- 0.841017      6        bad
chromium@443455                    9.00642 +- 0.698582      6        bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.news.nytimes v8.browsing_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8990305261781336384

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5872020971061248


| 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!

Regresses the scavenging times by 10%-20% (see https://chromeperf.appspot.com/group_report?bug_id=681541) real web benchmarks. This is pretty radical as this CL also makes it into M57.

Any short-term plans? Can we temporarily revert and fix this one offline?

I'd rather not, it's a large CL.
Working on a fix to another regression, I hope that it is this one as well.

Sorry to forget this: the CL was ultimately reverted, and the graph should have recovered.