All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=712657

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg7LWarwsM


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

chromium-rel-win7-gpu-intel

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8981968798258900272

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

Hi ehmaldonado@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 : ehmaldonado
  Commit : 78eee2ed541a59974558681e9165a18a0fe5028b
  Date   : Tue Mar 28 13:16:54 2017
  Subject: Use local pages for webrtc telemetry tests.

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : webrtc.peerconnection
  Metric       : cpu_utilization/cpu_utilization_gpu
  Change       : 58.13% | 3.48007889195 -> 1.45703681901

Revision             Result                   N
chromium@460054      3.48008 +- 1.37384       6      good
chromium@460072      3.39168 +- 0.738659      6      good
chromium@460077      3.12208 +- 0.60959       6      good
chromium@460078      1.66355 +- 1.09933       6      bad       <--
chromium@460079      1.28867 +- 0.642854      6      bad
chromium@460081      1.41423 +- 0.765364      6      bad
chromium@460089      1.9624 +- 1.07645        6      bad
chromium@460124      1.78325 +- 1.0281        6      bad
chromium@460193      1.45704 +- 0.665621      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 webrtc.peerconnection

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981968798258900272

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6472029181575168


| 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!

 Issue 711264  has been merged into this issue.

Ned: Is something like this expected?

If the page content are exactly the same, this is unexpected. Please work with the author of cpu_utilization metric to figure out what happened here. I am not an expert in webrtc code.

Who could I ask?
This is the metric we're using
https://cs.chromium.org/chromium/src/tools/perf/metrics/cpu.py

We are trying to get rid of that cpu metric & replace it with the TBM2 version. You should ask someone from webrtc team for help with understanding the different CPU usage.

Downgrading P2s that haven't been modified in more than 6 months, which have no component or owner.