Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8987477642552977488

Oops, wrong test. Seems like the speedometer test started passing again and some other tests started failing a few days ago.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8987475358687595136

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

Hi dfalcantara@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Test failure found with culprit

Suspected Commit
  Author : dfalcantara
  Commit : c9625f069b73578ab45f49be70a830d48f6141bc
  Date   : Tue Feb 07 23:56:25 2017
  Subject: [Omaha] Make the whole pipeline run each time

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : speedometer
  Metric       : AngularJS-TodoMVC/AngularJS-TodoMVC

Revision             Exit Code      N
chromium@448802      0 +- N/A       5      good
chromium@448803      1 +- N/A       5      bad       <--
chromium@448804      1 +- N/A       5      bad
chromium@448805      1 +- N/A       5      bad
chromium@448807      1 +- N/A       5      bad
chromium@448811      1 +- N/A       5      bad
chromium@448820      1 +- N/A       5      bad
chromium@448838      1 +- N/A       5      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests speedometer

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987477642552977488

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6395726382235648


| 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!

What does this test actually do?  The CL is supposed to be running things on a background thread, so I don't know how it'd affect whatever this is.

Moreover, that CL updates Omaha code, so I don't know why it'd be affecting page loads.

This is a JavaScript performance benchmark, but it's an integration test. It opens up the browser, loads a page, waits for it to calculate performance metrics, and closes the browser. The logs are pretty hard to read (https://luci-logdog.appspot.com/v/?s=chrome%2Fbb%2Fchromium.perf%2FAndroid_Nexus5X_Perf__3_%2F3350%2F%2B%2Frecipes%2Fsteps%2Fspeedometer%2F0%2Fstdout) but it looks like the browser is not starting properly on Android.

Come to think of it, this CL was reverted and relanded on Feb 10th because of a startup crash.  The tests probably fixed themselves after the revert.  I'd argue that this bug should probably be marked WontFix.

Yeah, I think the speedometer test the bisect bot is yelling at you for starting working again. My mistake running bisect on it.

However, bot is still having a bunch of problems with other metrics such as v8.mobile_infinite_scroll_tbmv2. Bisect currently running on those.

Removing you as owner.

=== BISECT JOB RESULTS ===
Test failure found with culprit

Suspected Commit
  Author : dfalcantara
  Commit : c9625f069b73578ab45f49be70a830d48f6141bc
  Date   : Tue Feb 07 23:56:25 2017
  Subject: [Omaha] Make the whole pipeline run each time

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : v8.mobile_infinite_scroll_tbmv2
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:allocated_by_malloc:effective_size_avg/pinterest

Revision             Exit Code      N
chromium@448802      0 +- N/A       6      good
chromium@448803      1 +- N/A       6      bad       <--
chromium@448804      1 +- N/A       6      bad
chromium@448805      1 +- N/A       6      bad
chromium@448807      1 +- N/A       6      bad
chromium@448811      1 +- N/A       6      bad
chromium@448820      1 +- N/A       6      bad
chromium@448838      1 +- N/A       6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=pinterest v8.mobile_infinite_scroll_tbmv2

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8987475358687595136

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5135368833204224


| 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!

This regression has been open for half a year. It's not very actionable and the regression has been in all Chrome user's hands for months.