All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=744744

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=f7fa98856661eb9801f283b36626ec8391c0a360ac5ecaaea43ca5ebb6873dd4


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

chromium-rel-mac-retina

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8973798200156248080

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : after_load:power_avg/multitab_misc/multitab_misc_typical24


To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=multitab.misc.typical24 system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8973798200156248080


For feedback, file a bug with component Speed>Bisection

The ref build changed a little bit, but then the signal starts becoming more noisy. Adding Charlie to see if this is something we can look into.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971623694851923952

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Error: INFRA_FAILURE

The bisect was able to narrow the range, you can try running with:
  good_revision: 8c1ea7c7b7fc0559e586b8e8744a18855758c3a1
  bad_revision : 2ca6ae0f8886b307b24f8ff5a269a5921d0c093f

If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : after_load:power_avg/multitab_misc/multitab_misc_typical24

Revision             Result                   N
chromium@486750      6.08068 +- 0.233281      5      good
chromium@486870      6.06634 +- 0.260769      6      good
chromium@486930      6.39238 +- 0.337659      6      good
chromium@486989      10.0179 +- 0.249568      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=multitab.misc.typical24 system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8971623694851923952


For feedback, file a bug with component Speed>Bisection

I'm not sure why that last bisect bisect pooped out right in the middle. Going to try to relaunch it with the narrowed range.

simonhatch@, do you have any idea what happened here?

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971280235385680448

I believe there's a 1 hour step limit, and for whatever reason one of the steps took longer than an hour (normal perf test runs took about 15 mins prior to that). Not clear what causes it to freeze like this, will block on crbug.com/664765

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Error: INFRA_FAILURE

The bisect was able to narrow the range, you can try running with:
  good_revision: 4e145f3c41d31bd16f10b5a2bd2dfc0c38a3d87c
  bad_revision : 1b46145c5ed1e7266c1beaa3338c19ff547933be

If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : after_load:power_avg/multitab_misc/multitab_misc_typical24

Revision             Result                   N
chromium@486930      6.56914 +- 1.14463       6      good
chromium@486945      6.33133 +- 0.246864      6      good
chromium@486953      6.46046 +- 0.446997      5      good
chromium@486960      10.1174 +- 0.566069      6      bad
chromium@486989      10.0315 +- 0.379159      5      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=multitab.misc.typical24 system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8971280235385680448


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8971170182904069744

Third time's the charm...

One problem seems to be that with  bug 754746 , where the BattOr agent is crashing. I'm not sure why it's taking so long to do so, though.

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : after_load:power_avg/multitab_misc/multitab_misc_typical24


To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=multitab.misc.typical24 system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8971170182904069744


For feedback, file a bug with component Speed>Bisection

(╯°□°)╯︵ ┻━┻

This is actually failing due to a bug that I already fixed (after the last failed bisect), so _hopefully_ if I kick off the bisect again, it'll pick up the new version of Telemetry and I won't have to see this frickin' failure again.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8970280912470798064

Bisects run with the version of telemetry pinned to the revision they're running at. So I think the bisect will keep failing.

+dtu is it easier to do better with pinpoint?

Ah, shucks. Never mind then :-( It's likely this is just going to keep failing, then. For some reason, I thought the Catapult and benchmark isolate was pinned to HEAD but the chrome version varied. Looks like I was wrong.

=== BISECT JOB RESULTS ===
Bisect failed for unknown reasons

Please contact the team (see below) and report the error.


Bisect Details
  Configuration: mac_retina_perf_bisect
  Benchmark    : system_health.common_desktop
  Metric       : after_load:power_avg/multitab_misc/multitab_misc_typical24


To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=multitab.misc.typical24 system_health.common_desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8970280912470798064


For feedback, file a bug with component Speed>Bisection

Looks like we won't be able to get to the bottom of this, then. Charlie, as benchmark owner, please reopen if you think it's worth investigating manually.

Yes! By default Pinpoint will use the catapult specified by chromium/src DEPS, and you can override it to pin it to a particular catapult revision.

We've also thought about things like running the job with a particular patch or commit cherry-picked over the failing range, but that's still in the future.