All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=708550

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg3Kn05woM,agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg3Kn05wgM


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

android-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983144916770651584

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

Hi ricea@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 : ricea
  Commit : 6dbcaaae9b0de50e03a7b11cb0506a8b4b42923f
  Date   : Tue Apr 04 06:24:11 2017
  Subject: Ship WritableStream

Bisect Details
  Configuration: android_nexus5X_perf_bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/background/after_http_www_baidu_com_s_word_google
  Change       : 0.93% | 4088834.66667 -> 4126724.0

Revision             Result                  N
chromium@461623      4088835 +- 323.514      6      good
chromium@461635      4091659 +- 15732.0      6      good
chromium@461641      4091756 +- 16085.6      6      good
chromium@461642      4088736 +- 441.38       6      good
chromium@461643      4129519 +- 16122.0      6      bad       <--
chromium@461644      4129340 +- 15963.4      6      bad
chromium@461647      4126724 +- 405.995      6      bad

Please refer to the following doc on diagnosing memory regressions:
  https://chromium.googlesource.com/chromium/src/+/master/docs/memory-infra/memory_benchmarks.md

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 memory.top_10_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8983144916770651584

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5342723621519360


| 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!

Weird. Does this benchmark run with or without --experimental-web-platform-features? CCing other team members.

Issue 708519 has been merged into this issue.

 Issue 708549  has been merged into this issue.

Is this only regressing on Android?

 Issue 708549  is a regression in the speed of appendChild(). This makes no sense whatsoever.

ChromiumPerf/android-webview-nexus5X/blink_perf.bindings / first-child, node-list-access and undefined-first-child haven't regressed.
https://chromeperf.appspot.com/report?sid=0e409091f1956efb41caa266b818baa25ce89750fd8e71bef91b7720521b15b9

 Issue 708551  has been merged into this issue.

 Issue 708551  is "thread_times.key_idle_power_cases", again on Android.

A lot of the memory tests like this are clank-specific AFAIK so would only alert on Android.

As far as experimental features, it depends.  Runtime enabled features, yes: https://www.chromium.org/blink/runtime-enabled-features

Finch trials, depends on if the bots are configured to run them: https://cs.chromium.org/chromium/src/testing/variations/fieldtrial_testing_config.json?dr

 Issue 708953  has been merged into this issue.

 Issue 708953  is another 300K v8 heap size increase, but this time on Windows! Yay!

Looks like a 40kib regression; we've updated our alerting thresholds to ignore regressions under 128kib