All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=768866

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=f8bc18f7139bf4f005d12f3112e307a39a1c32561d65d1e29b41056940992e69


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

android-nexus7v2

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8967375739389668864

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

Hi hubbe@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 : Fredrik Hubinette
  Commit : ff3b31782d552b03104a6d831c7530605b52b13f
  Date   : Fri Sep 22 21:11:16 2017
  Subject: avoid thread-jump in multibuffer_datasource::read()

Bisect Details
  Configuration: android_nexus7_perf_bisect
  Benchmark    : media.android.tough_video_cases
  Metric       : seek/tulip2.wav_seek_warm
  Change       : 36.93% | 25.7414285714 -> 35.3096428571

Revision             Result                  N
chromium@503822      25.7414 +- 40.004       14      good
chromium@503848      25.3328 +- 6.50776      9       good
chromium@503849      24.19 +- 1.28043        6       good
chromium@503850      33.64 +- 7.19034        6       bad       <--
chromium@503852      34.7025 +- 3.13367      6       bad
chromium@503855      33.4072 +- 17.4758      9       bad
chromium@503861      34.24 +- 2.65056        6       bad
chromium@503874      35.1024 +- 55.5914      21      bad
chromium@503924      35.3096 +- 6.16986      14      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.android.tough_video_cases

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8967375739389668864


For feedback, file a bug with component Speed>Bisection

Hi Hubbe, it looks like one of you changes have caused a jump in the warm seek time metric in the media benchmark. Could you please take a look to see if this is expected, and resolve this bug accordingly? Thanks!

I'm working on a CL that I think will resolve this.
Even if that's not the case, if this regression only happens on nexus7, I'm probably not going to worry too much about it. 

This bug is old and probably not worth fixing anymore.
The CL involved is a speedup most of the time, but can cause minor slowdowns for low-end devices, most of which are now too old to really matter.