All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=631152

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg6oDamgkM


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

chromium-rel-win7-gpu-ati

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9006139632699940384

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

Hi jsbell@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Change WTF::TemporaryChange to be an alias for base::AutoReset
Author  : jsbell
Commit description:
  
The behavior is the same, except that base::AutoReset takes a pointer
whereas WTF::TemporaryChange took a non-const ref, so callers get
sprinkled with ampersands. Since there's no benefit to a different
name, align the name as well, and clean up #include usage.

Inheriting from a typename alias confused MSVC, so the one instance
of that is changed to use a member instead.

Review-Url: https://codereview.chromium.org/2151933003
Cr-Commit-Position: refs/heads/master@{#406285}
Commit  : bb8a6045d4557a81857321fc436b2e72930de689
Date    : Tue Jul 19 16:05:09 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N   Good?
chromium@406284  71.1104  1.80391   12  good
chromium@406285  72.6975  1.2972    8   bad    <--
chromium@406286  72.7188  1.24728   12  bad
chromium@406287  72.0442  1.14816   12  bad
chromium@406290  72.1267  0.855154  12  bad
chromium@406295  72.6544  1.29126   8   bad
chromium@406305  73.3175  2.29123   8   bad
chromium@406326  73.6003  2.30437   18  bad

Bisect job ran on: winx64ati_perf_bisect
Bug ID: 631152

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/garden2_10s.ogv_seek_cold
Relative Change: 3.08%
Score: 98.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1444
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9006139632699940384


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5041757004234752

| 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!

The *_ref line on the graph spikes up at the same time - doesn't that indicate that it's not actually a regression but something like load on the machine?

Doesn't look like it could be my change, but I can't spot anything else suspicious. Then again, I know nothing about video. Feel free to try a revert if you think this is real.

There seems to be a slew of these, with non-overlapping ranges, and none of the other bisects have found anything interesting. I think the bisect is wrong, but I don't know what is actually going on.

Perf sheriff ping: reminder to follow up on possible performance issues