All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=609785

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgtMrYtQoM


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

linux-release

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


=== Bisection aborted ===
The bisect was aborted because The metric values for the initial "good" and "bad" revisions do not represent a clear regression.
Please contact the the team (see below) if you believe this is in error.

=== Warnings ===
The following warnings were raised by the bisect job:

 * Bisect failed to reproduce the regression with enough confidence.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@391270  49.9667  1.32887  18  good
chromium@391308  52.0111  3.29507  18  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 609785

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.pywebsocket
Test Metric: XHR-send-arraybuffer-window-async-verify/XHR-send-arraybuffer-window-async-verify
Relative Change: 0.08%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6474
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013418398012313536


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5230117077385216

| 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!

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

Hi pauljensen@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 : [Cronet] Resurrect Android USB reverse tethering
Author  : pauljensen
Commit description:
  
This was removed in crrev.com/1828703004 but is needed for Cronet
perf test.  This python script sets up USB reverse tethering for
an Android device, thereby allowing extremely fast networking
between the device and host computer.  Upon successful initialization
of USB reverse tethering, a network interface should be up on both
the device and host computer and provisioned with an IP address
allowing routing of packets between the device and host computer.

Review-Url: https://codereview.chromium.org/1942283002
Cr-Commit-Position: refs/heads/master@{#391280}
Commit  : ce7be43d95149a1a97a6b4088127fe0cfe895831
Date    : Tue May 03 17:31:37 2016


===== TESTED REVISIONS =====
Revision         Mean    Std Dev   N  Good?
chromium@391270  69.24   2.66889   5  good
chromium@391275  65.725  1.45774   8  good
chromium@391278  65.75   1.39591   8  good
chromium@391279  66.5    1.1726    5  good
chromium@391280  62.84   0.988939  5  bad    <--
chromium@391289  63.5    0.992472  5  bad
chromium@391308  65.65   1.74438   8  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 609785

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.pywebsocket
Test Metric: XHR-send-text-window-sync-verify/XHR-send-text-window-sync-verify
Relative Change: 5.34%
Score: 99.8

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6476
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013414825917291040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6368191941443584

| 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!

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

Hi anthonyvd@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 : Fix a used-after-free caused by an unremoved Observer.
Author  : anthonyvd
Commit description:
  
ProfileInfoHandler starts observing the ProfileInfoCache when it receives the
JavascriptAllowed event and stops on JavascriptDisallowed, which isn't always
called. This CL uses ScopedObserver to ensure ProfileInfoHandler instances are
never freed without removing themselves as Observers.

BUG= 607921 

Review-Url: https://codereview.chromium.org/1942323002
Cr-Commit-Position: refs/heads/master@{#391320}
Commit  : 7426febf7c71bba00d58289d4066753012a04d81
Date    : Tue May 03 19:12:26 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N   Good?
chromium@391250  61.7     0.667618  8   good
chromium@391285  62.1167  1.05816   12  good
chromium@391303  63.675   3.29604   12  good
chromium@391312  68.86    1.31072   5   good
chromium@391316  68.26    1.31833   5   good
chromium@391318  67.26    1.37768   5   good
chromium@391319  65.88    1.24378   5   good
chromium@391320  60.7875  0.412094  8   bad    <--

Bisect job ran on: linux_perf_bisect
Bug ID: 609785

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.pywebsocket
Test Metric: fetch-send-text-window-async-verify/fetch-send-text-window-async-verify
Relative Change: 1.59%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6475
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013414836955479424


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5813421417693184

| 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!

My change is irrelevant unless the experimental Material Design Settings page is opened, then closed. Even then, it just properly removes an Observer.

I fail to see how it could be related to what I understand is a blink websocket perf regression, so setting this to untriaged so it gets addressed.

Closing as WontFix, because:
From the graph https://chromeperf.appspot.com/group_report?bug_id=609785,
- looks recovered around 392135, and
- reference build was also regressing.
From bisect results,
- No clear culprit was found, and
- Regressions reported by bisect reports are <~10% (which I think we can safely ignored), not ~20%.