All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=692963

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg2PHmpwkM


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

chromium-rel-win7-gpu-ati

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8987514486317982896

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

Hi mustaq@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 : mustaq
  Commit : e3214ac5f080260c9e0584fc1776dad446ac33e9
  Date   : Wed Feb 08 23:46:56 2017
  Subject: Added a null check in WebPluginContainerImpl::handleMouseEvent.

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : blink_perf.canvas
  Metric       : draw-static-canvas-2d-to-hw-accelerated-canvas-2d/draw-static-canvas-2d-to-hw-accelerated-canvas-2d
  Change       : 13.57% | 259078.194066 -> 223917.981287

Revision             Result                 N
chromium@449025      259078 +- 23320.7      6      good
chromium@449096      241330 +- 13458.9      9      good
chromium@449132      249648 +- 14415.6      6      good
chromium@449137      252188 +- 13846.8      6      good
chromium@449139      258492 +- 8825.94      6      good
chromium@449140      245497 +- 6261.08      6      good
chromium@449141      230963 +- 5201.4       6      bad       <--
chromium@449150      220665 +- 4035.68      6      bad
chromium@449167      231321 +- 10154.5      6      bad
chromium@449309      223918 +- 10058.7      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 blink_perf.canvas

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987514486317982896

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5828956055404544


| 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!

Not sure I believe that bisect.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8987503965353601984

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

Hi rtoy@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 : rtoy
  Commit : c9ef493e1d238a655011a872886903a1d44d52bd
  Date   : Thu Feb 09 19:58:21 2017
  Subject: Continue to process AnalyserNode if inputs are silent.

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : blink_perf.canvas
  Metric       : draw-static-canvas-2d-to-hw-accelerated-canvas-2d/draw-static-canvas-2d-to-hw-accelerated-canvas-2d
  Change       : 12.31% | 256710.669134 -> 225100.546271

Revision             Result                 N
chromium@448500      256711 +- 12658.3      6       good
chromium@448950      241772 +- 19313.2      6       good
chromium@449175      259667 +- 13740.1      6       good
chromium@449288      263437 +- 12611.4      6       good
chromium@449344      264089 +- 17672.0      6       good
chromium@449372      265733 +- 14475.5      6       good
chromium@449376      256317 +- 23687.0      14      good
chromium@449378      252971 +- 30510.9      14      good
chromium@449379      243268 +- 23406.8      14      bad       <--
chromium@449386      230723 +- 34363.3      6       bad
chromium@449400      225101 +- 8889.58      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 blink_perf.canvas

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987503965353601984

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5130759527989248


| 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!

AFAICT, this test doesn't use an Analyser node (the linked to culprit).  The bisect seems wrong.

Reassigning back to alexclarke for further triage.

Explictly assigning. A CL you landed tripped one of the speed metrics we measure in the lab. If this is the first time this has happened to one of your CLs, or if it's been a while, please read: https://chromium.googlesource.com/chromium/src/+/master/docs/speed/addressing_performance_regressions.md

We're looking for one of the following:
1. Justification via explanation
2. Plan to revert or fix
3. Angry rage throwing of equipment at my head

Just be aware that I'm trained in trumpet playing and First Aid and am not afraid to use it.

Note: This was a bulk edit message and not very personal.

This needs retriaging, both the bisects in #c3 and #c6 seem obviously wrong.

junov, as owner of blink_perf.canvas, feel free to reopen if you think it's important, but I'm going to WontFix this as:

* The regression is from February
* Extending out the graph, There are huge spikes in performance from October-February but the data is super noisy, and it comes back down in this bug.
* Didn't happen on other test cases or bots.