All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=703237

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgjKqMogoM


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

chromium-rel-mac11

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984582567001734592

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

The bisect was able to narrow the range, you can try running with:
  good_revision: 1533215983ea3066ce9060cd6fdef87dc8f452ca
  bad_revision : 480869caaa149d728cbc2381476dbb5a153dc652

If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : media_perftests
  Metric       : audio_bus_to_interleaved/int8_t
  Change       : 38.42% | 20.9348416667 -> 28.9771666667

Revision             Result                   N
chromium@457278      20.9348 +- 0.181231      6      good
chromium@457293      21.0124 +- 0.34476       6      good
chromium@457301      20.9467 +- 0.171018      6      good
chromium@457303      28.9745 +- 0.243358      6      bad
chromium@457305      28.9423 +- 0.24087       6      bad
chromium@457308      28.9175 +- 0.199203      6      bad
chromium@457337      29.0964 +- 0.518101      6      bad
chromium@457396      28.9772 +- 0.26312       6      bad

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

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984582567001734592

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6330299174092800


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984579167763648752

Retrying with recommended revision range.

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

Hi tguilbert@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 : tguilbert
  Commit : cd31831bb240ce19d3e3af6f2b95f8ad11c96e5c
  Date   : Thu Mar 16 01:35:03 2017
  Subject: Fix unsupported audio channel layout

Bisect Details
  Configuration: mac_10_11_perf_bisect
  Benchmark    : media_perftests
  Metric       : audio_bus_to_interleaved/int8_t
  Change       : 38.86% | 20.912275 -> 29.0397

Revision             Result                   N
chromium@457301      20.9123 +- 0.170531      6      good
chromium@457302      29.16 +- 0.194864        6      bad       <--
chromium@457303      29.0397 +- 0.273259      6      bad

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

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8984579167763648752

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5804267954241536


| 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!

+crouleau

Do you think this might be a dupe of crbug/688610? The test still seems flaky, even if the bisect is very confident.

Yeah, I think we should probably delete this test because we don't care when it regresses. I think +Dale was planning on deleting this test.

Yeah, dupe away. I'll eventually get to deleting this.

I have another audio perf test that is supposedly regressed - audio_converter/convert_fifo_only. Do we care about this one?

Issue 703239

Yes; this one we don't care about because no one uses int8_t; though it's always surprising when it regresses.