All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=772955

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=c3f82cbdb159436a8abc63fc2d9560a9f9321d2672c1491e33d24f0564be4e07


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

chromium-rel-win7-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8966188637921503488

=== Auto-CCing suspected CL author danyao@google.com ===

Hi danyao@google.com, 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 : Danyao Wang
  Commit : 966a6107733d96f82ab9701ae2bf08ce18ff6511
  Date   : Mon Oct 02 18:44:05 2017
  Subject: [Nav Experiment] Implement GoBack() and GoForward() using GoToIndex().

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : media_perftests
  Metric       : audio_bus_to_interleaved/int32_t
  Change       : 20.60% | 20.4128833333 -> 16.2071583333

Revision             Result                    N
chromium@505696      20.4129 +- 0.446043       6      good
chromium@505697      17.4839 +- 0.241172       6      bad       <--
chromium@505698      18.0207 +- 0.293999       6      bad
chromium@505700      17.4808 +- 0.404428       6      bad
chromium@505703      18.2573 +- 1.19543        6      bad
chromium@505709      17.4855 +- 2.00273        6      bad
chromium@505721      16.2106 +- 0.109154       6      bad
chromium@505745      16.3377 +- 0.421809       6      bad
chromium@505793      16.2072 +- 0.0901173      6      bad

To Run This Test
  .\src\out\Release\media_perftests.exe --single-process-tests

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966188637921503488


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8966099471375026128

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

Suspected Commit
  Author : Danyao Wang
  Commit : 966a6107733d96f82ab9701ae2bf08ce18ff6511
  Date   : Mon Oct 02 18:44:05 2017
  Subject: [Nav Experiment] Implement GoBack() and GoForward() using GoToIndex().

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : media_perftests
  Metric       : audio_bus_to_interleaved/int32_t
  Change       : 22.08% | 20.2393388889 -> 16.2341666667

Revision             Result                    N
chromium@505696      20.2393 +- 3.10438        9      good
chromium@505697      18.7721 +- 4.69248        9      bad       <--
chromium@505698      18.3386 +- 0.730654       6      bad
chromium@505700      17.752 +- 0.898097        6      bad
chromium@505703      17.9625 +- 0.909411       6      bad
chromium@505709      16.7525 +- 1.26115        6      bad
chromium@505721      16.6646 +- 2.52997        6      bad
chromium@505745      16.1961 +- 0.0710339      6      bad
chromium@505793      16.2342 +- 0.0761236      6      bad

To Run This Test
  .\src\out\Release\media_perftests.exe --single-process-tests

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8966099471375026128


For feedback, file a bug with component Speed>Bisection

I'm not sure how my change could affect the audio_bus_to_interleaved metric in win_perf_biset, because the CL only affects iOS code and is currently gated behind an experiment flag...

However, running biset a second time reproduced the result of the first biset.
According to the bug report, audio_bus_to_interleaved has returned to "normal" value (~18) from the elevated value (~20), and my change has not been reverted. Could there be another change that happened in that time frame?

Since the benchmark has reverted to the previous values, marking this as done.