All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=607744

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg5OO0twoM


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

android-nexus5X

Previous bisect job failed in provision devices step; trying again: https://chromeperf.appspot.com/buildbucket_job_status/9013387576280128816

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

Hi bmeurer@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 : [turbofan] Enable concurrent (re)compilation.
Author  : bmeurer
Commit description:
  
Refactor the TurboFan pipeline to allow for concurrent recompilation in
the same way that Crankshaft does it. For now we limit the concurrent
phases to scheduling, instruction selection, register allocation and
jump threading.

R=mstarzinger@chromium.org, ahaas@chromium.org, jarin@chromium.org

Review URL: https://codereview.chromium.org/1179393008

Cr-Commit-Position: refs/heads/master@{#35818}
Commit  : ff19726d8045a11df0535d8cb9a51e8b7a35c5cc
Date    : Wed Apr 27 12:40:00 2016


===== TESTED REVISIONS =====
Revision                       Mean      Std Dev    N  Good?
chromium@390093                0.231037  0.0427808  8  good
chromium@390093,v8@7f3954c57b  0.250035  0.0324224  8  good
chromium@390093,v8@ff19726d80  0.336965  0.0487011  5  bad    <--
chromium@390093,v8@42c0e2ec7c  0.340637  0.048834   8  bad
chromium@390094                0.367442  0.0476806  5  bad
chromium@390095                0.367906  0.0406768  5  bad
chromium@390097                0.388785  0.0329269  5  bad
chromium@390100                0.373552  0.0218307  5  bad
chromium@390106                0.38874   0.0211884  5  bad
chromium@390118                0.355022  0.0449406  8  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 607744

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_mobile_sites_smooth
Test Metric: thread_other_cpu_time_per_frame/thread_other_cpu_time_per_frame
Relative Change: 47.64%
Score: 99.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/142
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013387576280128816


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5339383192879104

| 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!

All Supported for online program 

bmeurer@ do you have any update on this regression?

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

bmeurer: ping?

cc-ing test owner vmiura

Emailed bmeurer.

Sorry, I missed this one. Can someone explain to me what this graph/benchmark is about? I don't see how this is related to my CL, especially since there's already a spike before my CL.

Ah, got it, this measures the CPU consumption of the non-main thread CPUs, which is obviously higher if we move work from the main thread to a separate thread, as done by this CL. So this is then of course working as intended. :-)

Hmmm... are you sure? It doesn't look like main thread CPU time went down, or any other thread for that matter.

https://chromeperf.appspot.com/report?sid=e96d9ca26286b0a3042efdff71f52ee2aca4579b028a6bd2f7a877e40ae1b68e&rev=390503

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9007320838530283760

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

Hi rockot@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 : Add field trial testing config for MojoChannel
Author  : rockot
Commit description:
  
R=asvitkine@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1918583002

Cr-Commit-Position: refs/heads/master@{#389497}
Commit  : bf36f2d46f43eff11f56ea205615f092801b1004
Date    : Mon Apr 25 16:47:03 2016


===== TESTED REVISIONS =====
Revision         Mean      Std Dev     N  Good?
chromium@389398  0.964412  0.0145936   8  good
chromium@389467  0.977241  0.0110447   8  good
chromium@389484  0.967996  0.0187807   8  good
chromium@389493  0.962924  0.0179442   5  good
chromium@389495  0.97154   0.0248619   8  good
chromium@389496  0.987973  0.0172194   8  good
chromium@389497  1.00973   0.00660377  8  bad    <--
chromium@389501  1.00093   0.0126174   8  bad
chromium@389535  1.00921   0.0173717   5  bad
chromium@389672  1.01396   0.0166722   8  bad
chromium@389946  1.0168    0.00444024  5  bad
chromium@390503  1.01577   0.0184271   5  bad

Bisect job ran on: android_nexus9_perf_bisect
Bug ID: 607744

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_mobile_sites_smooth
Test Metric: thread_renderer_compositor_cpu_time_per_frame/thread_renderer_compositor_cpu_time_per_frame
Relative Change: 5.08%
Score: 99.5

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus9_perf_bisect/builds/1886
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9007320838530283760


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5857970156470272

| 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!

If that experiment increased CPU time it wouldn't be terribly surprising, but we've since:

 - reduced the CPU cost of the then-experimental code path
 - removed use of the experiment altogether (months ago)

So assuming the bisect is right this time, it's still WontFix.