All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=844337

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=4cdc8534f41aab77b75b3b7a8b1764155a24517e31e84cd72b5b0ef3a8193eb3


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

health-plan-webview-low-end-phone
perf-go-phone-1024
perf-go-webview-phone

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8946207432455069472

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

Bisect Details
  Configuration: go-phone-1024-perf-bisect
  Benchmark    : system_health.memory_mobile
  Metric       : memory:chrome:all_processes:reported_by_chrome:v8:effective_size_avg/load_games/load_games_spychase
  Change       : 0.32% | 7057884.0 -> 7035356.0

Revision                       Result                 N
android-chrome@4426288b2b      7057884 +- 107224      21      good
android-chrome@dda312c828      7035356 +- 164310      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-chrome --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=load.games.spychase 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/8946207432455069472


For feedback, file a bug with component Speed>Bisection

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8945848446938971328

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8945848436405439360

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Please try rerunning the bisect.


If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: webview-low-end-phone-perf-bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/foreground/http_www_baidu_com_s_word_google
  Change       : 5.96% | 2984694.66667 -> 3162650.0

Revision                                                     Result                  N
android-chrome@4426288b2b                                    2984695 +- 21463.3      6      good
android-chrome@4426288b2b,chromium@558065                    3002994 +- 7969.26      6      good
android-chrome@4426288b2b,chromium@558262                    2998032 +- 11560.9      6      good
android-chrome@4426288b2b,chromium@558311                    2987853 +- 14010.4      6      good
android-chrome@4426288b2b,chromium@558313                    3001258 +- 10422.5      6      good
android-chrome@4426288b2b,chromium@558313,v8@bf05c62777      3162417 +- 290.857      6      bad
android-chrome@4426288b2b,chromium@558314                    3162221 +- 500.773      6      bad
android-chrome@4426288b2b,chromium@558315                    3161817 +- 847.666      6      bad
android-chrome@4426288b2b,chromium@558318                    3162145 +- 752.351      6      bad
android-chrome@4426288b2b,chromium@558324                    3162459 +- 242.12       6      bad
android-chrome@4426288b2b,chromium@558336                    3135893 +- 64563.3      6      bad
android-chrome@4426288b2b,chromium@558360                    3162085 +- 686.457      6      bad
android-chrome@dda312c828                                    3162026 +- 651.731      6      bad
android-chrome@33f296deb5                                    3162650 +- 261.378      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.www.baidu.com.s.word.google memory.top_10_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8945848446938971328


For feedback, file a bug with component Speed>Bisection

=== BISECT JOB RESULTS ===
Bisect was unable to run to completion

Please try rerunning the bisect.


If failures persist contact the team (see below) and report the error.


Bisect Details
  Configuration: go-webview-phone-perf-bisect
  Benchmark    : memory.top_10_mobile
  Metric       : memory:webview:all_processes:reported_by_chrome:v8:effective_size_avg/foreground/http_www_baidu_com_s_word_google
  Change       : 5.94% | 2985322.66667 -> 3162514.0

Revision                                                     Result                  N
android-chrome@4426288b2b                                    2985323 +- 8948.81      6      good
android-chrome@4426288b2b,chromium@558065                    2992792 +- 9493.87      6      good
android-chrome@4426288b2b,chromium@558262                    2987163 +- 8616.27      6      good
android-chrome@4426288b2b,chromium@558311                    2992707 +- 10584.0      6      good
android-chrome@4426288b2b,chromium@558313                    2990585 +- 11884.5      6      good
android-chrome@4426288b2b,chromium@558313,v8@bf05c62777      3142916 +- 58252.6      9      bad
android-chrome@4426288b2b,chromium@558314                    3162159 +- 722.421      6      bad
android-chrome@4426288b2b,chromium@558318                    3162061 +- 482.523      6      bad
android-chrome@4426288b2b,chromium@558324                    3162383 +- 292.451      6      bad
android-chrome@4426288b2b,chromium@558336                    3162041 +- 606.829      6      bad
android-chrome@4426288b2b,chromium@558360                    3162296 +- 104.827      6      bad
android-chrome@dda312c828                                    3162514 +- 125.647      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-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http.www.baidu.com.s.word.google memory.top_10_mobile

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

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8945848436405439360


For feedback, file a bug with component Speed>Bisection

Given the previous partial bisects:

https://chromium.googlesource.com/v8/v8/+log/162ca8aa..d936a701

bad -> bf05c62 [turbofan] Check the return value of AssembleDeoptimizerCall. by Georgia Kouveli · 9 days ago
dba31f6 [wasm] Only use retpoline if untrusted code mitigations are enabled by Clemens Hammacher · 9 days ago
db4f1ee [lsan] Disable annotations on windows by Clemens Hammacher · 9 days ago
0a71347 Revert "[tools] Add benchmark owners to the config" by Sergiy Byelozyorov · 9 days ago
91ddb65 Revert promises optimizations due to regressions in async hooks by Maya Lekova · 9 days ago
good -> [tools] Add benchmark owners to the config

The most likely culprit is

91ddb65 Revert promises optimizations due to regressions in async hooks by Maya Lekova · 9 days ago

Maya, could you please take a look?

Looks like the regression is in old space: https://chromeperf.appspot.com/report?sid=b27b9388e8166f69435634dc0ddfed55565d1a4389ed301718a505d7aabc9478&start_rev=1524460402&end_rev=1526989787

This revert gets back an additional promise, which had previously been optimized away. Unfortunately this optimization caused a regression in the async hooks validity in Node.js (see https://github.com/nodejs/node/issues/20516), so landing it back will take some time.
Benedikt, can you provide your insight as well?

Yeah this is expected, since we undid a couple of optimizations that also save memory. We'll get it back eventually.