All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=699929

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDglLfo7wgM


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

chromium-rel-win7-x64-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8985611075659711952

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

Hi binji@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 : binji
  Commit : cb8fb46aa3f1bc785f89bc16ad389f7a9b4fe982
  Date   : Tue Feb 28 20:31:02 2017
  Subject: Move ArrayBuffer.prototype.slice implementation to C++

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : v8.runtimestats.browsing_desktop
  Metric       : V8 C++:duration_avg/browse_media/browse_media_tumblr
  Change       : 8.77% | 310.432833333 -> 337.665

Revision                           Result                  N
chromium@453725                    310.433 +- 8.98414      6      good
chromium@453756                    312.58 +- 5.2745        6      good
chromium@453764                    310.991 +- 8.82174      6      good
chromium@453764,v8@ae66dcbec0      311.668 +- 8.96581      6      good
chromium@453764,v8@9c385f0405      310.992 +- 6.02386      6      good
chromium@453764,v8@cb8fb46aa3      335.423 +- 6.39332      6      bad       <--
chromium@453764,v8@03d03326a5      331.489 +- 9.40785      6      bad
chromium@453765                    335.018 +- 5.98446      6      bad
chromium@453766                    334.456 +- 9.9882       6      bad
chromium@453768                    336.681 +- 5.17026      6      bad
chromium@453771                    336.756 +- 5.25049      6      bad
chromium@453786                    336.133 +- 9.07599      6      bad
chromium@453847                    337.665 +- 7.08569      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=browse.media.tumblr v8.runtimestats.browsing_desktop

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8985611075659711952

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4967222197354496


| 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!

Looks like this was subsequently improved by the switch to TF+I in v8.

I think it makes sense that more time is spent in C++ when you move more code into C++. I suppose this is as expected? Feel free to WontFix in that case.