All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=736771

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgzvyD7AgM


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

chromium-rel-mac11-pro

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8975712122696604512

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

Hi danno@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 : danno
  Commit : e2544f6c03cb6725328036192a64fd80fad0ff1d
  Date   : Thu Jun 22 15:43:35 2017
  Subject: Fix deoptmization of inlined TF instanceOf to call ToBoolean

Bisect Details
  Configuration: mac_pro_perf_bisect
  Benchmark    : media.tough_video_cases_tbmv2
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/video.html?src_garden2_10s.mp4_seek
  Change       : 1.23% | 4576531.55556 -> 4632647.11111

Revision                           Result                  N
chromium@481654                    4576532 +- 495486       9       good
chromium@481733                    4632395 +- 129.326      6       good
chromium@481761                    4632465 +- 770.699      9       good
chromium@481768                    4545343 +- 476517       6       good
chromium@481769                    4632452 +- 782.62       9       good
chromium@481769,v8@731d1b73af      4632376 +- 137.171      6       good
chromium@481769,v8@e2544f6c03      4632622 +- 166.96       9       bad       <--
chromium@481769,v8@14ba446bf2      4632622 +- 166.96       9       bad
chromium@481770                    4632740 +- 750.616      6       bad
chromium@481772                    4632796 +- 679.176      6       bad
chromium@481775                    4595406 +- 502161       14      bad
chromium@481789                    4632647 +- 139.689      9       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=video.html.src.garden2.10s.mp4.seek media.tough_video_cases_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8975712122696604512

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5270005769306112


| 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!

 Issue 737491  has been merged into this issue.

This was a correctness fix, and the "regression" often seem to be graphs shifting to tending to stay at the worse end of a bimodal state of a benchmark. Going to keep an eye on this over time to see if this is just a shift in the result's natural bias or something real.

Seems like this is WontFix.