All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=726948

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-rzjugoM


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

android-nexus5

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8978488153643145328

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_social/load_social_pinterest

Revision             Result                  N
chromium@473422      54388536 +- 786408      21      good
chromium@473434      54368397 +- 979930      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-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.social.pinterest system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978488153643145328

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6384839315423232


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8978484090103740464

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_social/load_social_pinterest

Revision             Result                  N
chromium@473422      54496348 +- 942778      21      good
chromium@473434      54529019 +- 798316      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-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.social.pinterest system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978484090103740464

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6384839315423232


| 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!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8976998468904035344

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

Hi neis@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 : Georg Neis
  Commit : 233b6926e3ae29a4567b4c55bc1c1e96c07c1f92
  Date   : Fri May 19 13:15:25 2017
  Subject: [compiler] Generate all deopt entries upfront.

Bisect Details
  Configuration: android_nexus5_perf_bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_os:system_memory:native_heap:proportional_resident_size_avg/load_social/load_social_pinterest
  Change       : 0.61% | 54122393.3333 -> 54450926.6667

Revision                           Result                  N
chromium@473200                    54122393 +- 291790      6      good
chromium@473238                    53993199 +- 485174      6      good
chromium@473257                    54062319 +- 381882      9      good
chromium@473257,v8@34c1c76d8a      53982674 +- 542222      9      good
chromium@473257,v8@233b6926e3      54579268 +- 652541      6      bad       <--
chromium@473257,v8@ee83890103      54486937 +- 472821      6      bad
chromium@473257,v8@a621462bab      54351599 +- 344059      6      bad
chromium@473258                    54453828 +- 555907      6      bad
chromium@473259                    54491204 +- 511237      6      bad
chromium@473260                    54373387 +- 609589      9      bad
chromium@473262                    54429138 +- 401062      9      bad
chromium@473266                    54408203 +- 542967      9      bad
chromium@473275                    54399670 +- 322976      9      bad
chromium@473350                    54500932 +- 441561      6      bad
chromium@473500                    54450927 +- 295524      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 --story-filter=load.social.pinterest system_health.memory_mobile

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8976998468904035344

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6616727242145792


| 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!

This change is necessary in order to move Turbofan code generation off the main thread (which is expected to benefit performance).