All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=613893

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICgzNeUpgoM


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

android-nexus5X

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

Hi dominicc@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 a tuple of name, local name for hashing custom element definitions.
Author  : dominicc
Commit description:
  
BUG= 594918 

Review-Url: https://codereview.chromium.org/1996213002
Cr-Commit-Position: refs/heads/master@{#395254}
Commit  : 4de05fac9bf71bd0724f7a0becb0668cf43ab763
Date    : Sat May 21 07:42:34 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@395253  232.273  0.92404   5  good
chromium@395254  234.817  0.953589  5  bad    <--
chromium@395255  234.892  1.10686   5  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 613893

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.css
Test Metric: ClassDescendantSelector/ClassDescendantSelector
Relative Change: 1.13%
Score: 99.5

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/182
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9011901962980577680


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5839277976453120

| 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!

Dominic, seems your patch caused a regression. Could you take a look?

Hmm, we haven't shipped custom elements v1 yet, so this having a performance impact is a bit of a surprise.

Can you help me interpret these results? In particular I see a graph that goes down, and then up--see the attached snippet. Is there still a problem here?

Deleted: Screenshot from 2016-07-04 10:39:42.png 7.7 KB

	Screenshot from 2016-07-04 10:39:42.png 7.7 KB View Download

Moving this nonessential bug to the next milestone.

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

dominicc@: it's possible that your work cause a regression, and someone else's work cause an improvement (if you didn't revert your CL).

In this theoretical scenario, we are still losing a performance win due to your CL causing a regression. Can you investigate if this is the case?

pinging dominicc@

Yes that is possible but it seems unlikely--isn't it a remarkable coincidence that the "after" performance is in line with the "before" performance?

Could you rerun the bisect?

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005516415545499040

Lets try another bisect and see what it finds.

===== BISECT JOB RESULTS =====
Status: failed


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N   Good?
chromium@395299  235.29   0.534901  12  good
chromium@395320  235.882  0.744885  12  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 613893

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.css
Test Metric: ClassDescendantSelector/ClassDescendantSelector
Relative Change: 0.22%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/411
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9005516415545499040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5858349162168320

| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005310919444541648

===== BISECT JOB RESULTS =====
Status: failed


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N   Good?
chromium@395299  235.29   0.534901  12  good
chromium@395320  235.882  0.744885  12  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 613893

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.css
Test Metric: ClassDescendantSelector/ClassDescendantSelector
Relative Change: 0.22%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/411
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9005516415545499040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5858349162168320

| 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!

Could you help me understand these results? It seems there's a number that is slightly higher (eg 235.29~>235.882 = 235.29 * 1.0025) but what does status: "failed" mean?

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9005219733342871504

===== 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@395253  225.853  4.69201  18  good
chromium@395255  225.276  4.63201  18  bad

Bisect job ran on: android_nexus5X_perf_bisect
Bug ID: 613893

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.css
Test Metric: ClassDescendantSelector/ClassDescendantSelector
Relative Change: 0.87%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5X_perf_bisect/builds/437
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9005516415545499040


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5858349162168320

| 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!

Shall we just wontfix this? I think it was a blip.

Yeah, sorry about this, if you look at the bisect in #2 it only found a 1% change. Not sure what happened with this test but it's unlikely to be worth investigating.