All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=787202

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=0957cfe69935b96d63896f1f009f9aec18e73c919a0ba4a45ef99ca8c8f70c03


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

android-webview-nexus5X
android-webview-nexus6

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8962359672621955264

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

Hi leszeks@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 : Leszek Swirski
  Commit : 5aa6f58952047bf3eb72e8a016ae1520b62176fe
  Date   : Fri Nov 17 13:14:21 2017
  Subject: [cleanup] Move preparsed data into function data in SFI

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/load_media/load_media_facebook_photos
  Change       : 3.69% | 3865007.80952 -> 4007476.95238

Revision                           Result                  N
chromium@517228                    3865008 +- 138071       21      good
chromium@517470                    3866098 +- 130581       21      good
chromium@517531                    3871321 +- 69536.8      9       good
chromium@517546                    3866736 +- 102662       14      good
chromium@517550                    3870138 +- 83256.2      9       good
chromium@517552                    3855618 +- 62030.2      6       good
chromium@517553                    3860428 +- 76585.8      9       good
chromium@517553,v8@acfef3ec93      3863916 +- 82725.2      9       good
chromium@517553,v8@5aa6f58952      4035436 +- 386324       9       bad       <--
chromium@517553,v8@372fc681be      4009783 +- 210311       6       bad
chromium@517554                    3998754 +- 569691       21      bad
chromium@517561                    4029661 +- 426757       14      bad
chromium@517591                    4022245 +- 391400       9       bad
chromium@517711                    4007477 +- 654075       21      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.media.facebook.photos system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8962359672621955264


For feedback, file a bug with component Speed>Bisection

Either not this CL, or a WAI GC timing change -- this CL should have dropped memory use, not increased it (it decreased the size of one of the V8 classes by a pointer). It looks like the values in the bisect are within error, can we re-run it?

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8962293832478337408

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8962293811902085344

=== BISECT JOB RESULTS ===
Perf regression found but unable to continue

Bisect was stopped because a commit couldn't be classified as either
good or bad.


Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/load_media/load_media_facebook_photos

Revision             Result                 N
chromium@517200      3862821 +- 128184      21      good
chromium@517456      3874767 +- 148311      21      unknown
chromium@517711      3981152 +- 630492      21      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.media.facebook.photos system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8962293811902085344


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8962228886899025296

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

Suspected Commit
  Author : Leszek Swirski
  Commit : 5aa6f58952047bf3eb72e8a016ae1520b62176fe
  Date   : Fri Nov 17 13:14:21 2017
  Subject: [cleanup] Move preparsed data into function data in SFI

Bisect Details
  Configuration: android_webview_nexus6_aosp_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/load_media/load_media_facebook_photos
  Change       : 3.70% | 3865617.33333 -> 4008799.2381

Revision                           Result                  N
chromium@517228                    3865617 +- 155574       21      good
chromium@517470                    3882695 +- 127724       14      good
chromium@517531                    3864152 +- 88519.7      6       good
chromium@517546                    3864712 +- 75237.4      9       good
chromium@517550                    3847388 +- 79160.1      9       good
chromium@517552                    3867137 +- 116564       14      good
chromium@517553                    3860148 +- 83773.6      9       good
chromium@517553,v8@acfef3ec93      3856153 +- 103445       14      good
chromium@517553,v8@5aa6f58952      3993502 +- 590686       21      bad       <--
chromium@517553,v8@372fc681be      3971426 +- 407059       14      bad
chromium@517554                    3983364 +- 326049       9       bad
chromium@517561                    4088733 +- 313187       9       bad
chromium@517591                    3998996 +- 525070       21      bad
chromium@517711                    4008799 +- 632491       21      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.media.facebook.photos system_health.memory_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8962228886899025296


For feedback, file a bug with component Speed>Bisection

Well damn. +verwaest who suggested this change, any insight why shrinking SFIs could increase memory use?

It's measuring average usage; which is useless for measuring actual live bytes. We almost always see average increase somewhere when we clearly reduce memory usage.