All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=760160

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=391b0bf18fa690e4cebb0f0c442900344ae6f4349fe9e151190d3228156ee340


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

chromium-rel-win7-dual

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8969912371331012832

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

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


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Xida Chen
  Commit : 194b86fb0c8470eb7516e1f38eca1862166174c1
  Date   : Sat Aug 19 04:00:15 2017
  Subject: Add a Blink feature to turn off compositing animations

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : blink_perf.dom
  Metric       : remove_child_with_selection/remove_child_with_selection
  Change       : 3.57% | 1.77866666667 -> 1.84216666667

Revision             Result                    N
chromium@495791      1.77867 +- 0.0300888      6      good
chromium@495796      1.7715 +- 0.0185876       6      good
chromium@495798      1.75833 +- 0.0647457      9      good
chromium@495799      1.75917 +- 0.0318564      6      good
chromium@495800      1.81278 +- 0.0288367      9      bad       <--
chromium@495808      1.84217 +- 0.0687229      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 blink_perf.dom

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8969912371331012832


For feedback, file a bug with component Speed>Bisection

xidachen: any update on this bug?

This regression is due to an experiment that we want to conduct, which is outlined in crbug.com/754471. However, in order to best evaluate the experiment, we need to add more UMA metrics, and I have a WIP CL for that:
https://chromium-review.googlesource.com/c/chromium/src/+/636305

So I think the right thing to do here is block this on the experiment bug? 

Also +jbroman: do we want to have experimental features turned on for blink_perf.dom?

This feature doesn't seem like it should be marked experimental (and looking at this CL, it doesn't seem to be), so it shouldn't be on in the perf test.

I would expect blink_perf.dom to run in the "stable" configuration, as that's most representative of what we expect to ship to users.

Xida, IIUC the CL should have no behaviour change (unless something turns the feature on), correct? Do we expect this function to be so hot that the extra code slows this down?

jbroman@: I think you are right, I have not yet landed a CL to enable waterfall perf testing, so that should not have any behavior change.

A bad bisect?

Is this turned on as an experimental web platform feature? Note that they are enabled in blink_perf tests, although maybe they shouldn't be, at least for blink_perf.dom?

https://cs.chromium.org/chromium/src/tools/perf/benchmarks/blink_perf.py?q=blink_perf.py&sq=package:chromium&l=241

In the bisected CL, the change to the RuntimeEnabledFeatures config is:

+    {
+      name: "TurnOff2DAndOpacityCompositorAnimations",
+    },

Which does not mark the feature as experimental. Omitted status should not get caught by the "experimental" bucket, but should only be on if it's explicitly turned on by something.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965822582478740336

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965822574767708496

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8965822560035415088

I kicked off a few more bisects, we'll see what they come back with.

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Nico Weber
  Commit : 7cd2f6bd73c6a9c6200704e995b18ebb3168ead6
  Date   : Sat Aug 19 04:23:54 2017
  Subject: win: Turn off clang-by-default again for now.

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : blink_perf.dom
  Metric       : remove_child_with_selection/remove_child_with_selection
  Change       : 2.74% | 1.78533333333 -> 1.83422222222

Revision             Result                    N
chromium@495791      1.78533 +- 0.0700714      9      good
chromium@495800      1.79933 +- 0.0425598      6      good
chromium@495801      1.80167 +- 0.072535       6      good
chromium@495802      1.8885 +- 0.041491        6      bad       <--
chromium@495804      1.932 +- 0.132891         6      bad
chromium@495808      1.83422 +- 0.045095       9      bad

Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.dom

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965822574767708496


For feedback, file a bug with component Speed>Bisection

=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Nico Weber
  Commit : 7cd2f6bd73c6a9c6200704e995b18ebb3168ead6
  Date   : Sat Aug 19 04:23:54 2017
  Subject: win: Turn off clang-by-default again for now.

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : blink_perf.dom
  Metric       : textarea-edit/textarea-edit
  Change       : 11.03% | 2.88500488111 -> 2.56677246634

Revision             Result                    N
chromium@495791      2.885 +- 0.0220861        6      good
chromium@495800      2.87832 +- 0.0139956      6      good
chromium@495801      2.87547 +- 0.0197388      6      good
chromium@495802      2.56651 +- 0.0187586      6      bad       <--
chromium@495804      2.5649 +- 0.0206304       6      bad
chromium@495808      2.56677 +- 0.0188679      6      bad

Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.dom

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965822582478740336


For feedback, file a bug with component Speed>Bisection

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

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


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Michael Lippautz
  Commit : 146d898b6e27a3900dae50b497e24674d7d2adfb
  Date   : Mon Aug 21 09:24:08 2017
  Subject: [wrapper-tracing] NodeListsNodeData Align Oilpan and wrapper tracing

Bisect Details
  Configuration: win_perf_bisect
  Benchmark    : blink_perf.dom
  Metric       : remove_child_with_selection/remove_child_with_selection
  Change       : 19.42% | 1.767 -> 2.11016666667

Revision             Result                    N
chromium@495502      1.767 +- 0.0373363        6      good
chromium@495741      1.78367 +- 0.0147422      6      good
chromium@495860      1.84417 +- 0.0212328      6      good
chromium@495890      1.87317 +- 0.118612       6      good
chromium@495891      2.03333 +- 0.0197315      6      bad       <--
chromium@495892      2.0545 +- 0.0629087       6      bad
chromium@495894      2.059 +- 0.0666333        6      bad
chromium@495898      2.06367 +- 0.0377271      6      bad
chromium@495905      2.11683 +- 0.12334        6      bad
chromium@495920      2.05367 +- 0.0675376      6      bad
chromium@495979      2.11017 +- 0.0300472      6      bad

Please refer to the following doc on diagnosing blink_perf regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/speed/benchmark_harnesses/blink_perf.md

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.dom

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8965822560035415088


For feedback, file a bug with component Speed>Bisection

 Issue 760174  has been merged into this issue.

 Issue 760174  has been merged into this issue.