All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=690671

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgqLu7uggM


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

android-webview-nexus5X

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8988096623672297344

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

Hi yangguo@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 : yangguo
  Commit : c78d7fa1aefb4544ace6b248fe6ac247bf5cbed3
  Date   : Wed Feb 08 08:33:33 2017
  Subject: Link type feedback vectors to the shared function info.

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : thread_times.key_idle_power_cases
  Metric       : tasks_per_second_total_all/blank.html
  Change       : 8.67% | 2.8787230007 -> 3.12839356862

Revision                           Result                    N
chromium@448969                    2.87872 +- 0.090467       6      good
chromium@448970                    2.854 +- 0.163213         6      good
chromium@448970,v8@c78d7fa1ae      3.12048 +- 0.0596883      6      bad       <--
chromium@448970,v8@f0034287bf      3.14543 +- 0.158721       6      bad
chromium@448971                    3.13628 +- 0.0840113      6      bad
chromium@448972                    3.18659 +- 0.108195       6      bad
chromium@448974                    3.1535 +- 0.0476827       6      bad
chromium@448978                    3.16971 +- 0.117267       6      bad
chromium@448987                    3.12839 +- 0.0919695      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=blank.html thread_times.key_idle_power_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8988096623672297344

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5883077944934400


| 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!

@yangguO: Any idea why that patch would increase the number of tasks run for a blank tab?

very unlikely. the change simply adds one indirection in a data structure internal to v8

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8988008712020591136

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

Suspected Commit
  Author : yangguo
  Commit : c78d7fa1aefb4544ace6b248fe6ac247bf5cbed3
  Date   : Wed Feb 08 08:33:33 2017
  Subject: Link type feedback vectors to the shared function info.

Bisect Details
  Configuration: android_webview_arm64_aosp_perf_bisect
  Benchmark    : thread_times.key_idle_power_cases
  Metric       : tasks_per_second_total_all/blank.html
  Change       : 6.81% | 2.92079669277 -> 3.11970206433

Revision                           Result                    N
chromium@448969                    2.9208 +- 0.135812        6      good
chromium@448970                    2.89525 +- 0.100016       6      good
chromium@448970,v8@c78d7fa1ae      3.12804 +- 0.207728       6      bad       <--
chromium@448970,v8@f0034287bf      3.18627 +- 0.130861       6      bad
chromium@448971                    3.20327 +- 0.110324       6      bad
chromium@448972                    3.14467 +- 0.0702601      6      bad
chromium@448974                    3.13653 +- 0.108885       6      bad
chromium@448978                    3.1367 +- 0.109882        6      bad
chromium@448987                    3.1197 +- 0.153552        6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=blank.html thread_times.key_idle_power_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8988008712020591136

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5530593468612608


| 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!

It's been a couple months, and this regression looks real. yangguo - can you please either justify your CL or scope some work to address the regression you've included in your CL?

It's been over 6 months, looks like this will not get fixed.