All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=621173

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICggqTipQoM


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

chromium-rel-win7-x64-dual

Kicked off a bisect in https://chromeperf.appspot.com/buildbucket_job_status/9008940573191197680

Trying another bisect.

Moving this nonessential bug to the next milestone.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

YAB: https://chromeperf.appspot.com/buildbucket_job_status/9007409998933670480

 ===== TESTED REVISIONS ===== Revision Mean Std Dev N Good? chromium@399920 3.40833 0.0900337 12 good chromium@399951 3.46111 0.0832352 18 good chromium@399962 3.42222 0.095828 18 good chromium@399965 3.43889 0.0900254 18 good chromium@399968 3.53889 0.136722 18 bad chromium@399973 3.52222 0.0669113 18 bad

Oops, layout was broken.

Let me re-paste important lines.
chromium@399965 3.43889 0.0900254 18 good
chromium@399968 3.53889 0.136722 18 bad

In this range, only one suspicious CL was v8 roll.
https://codereview.chromium.org/2070593002

Let me assign hablich@ who was one of reviewers for this roll.

Assigning to this weeks memory sheriff.

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9004061586991020800

I'm not totally confident that this was a real regression in the first place, since this alert is only on one platform and the test is a bit noisy.

hpayer, if you think this is worth investigating, then I think the next step is to look through that v8 roll for suspicious CLs.

But I think it's likely not worth it; in that case, perhaps we should WontFix this bug.

===== BISECT JOB RESULTS =====
Status: failed


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev    N   Good?
chromium@399929                3.425    0.138873   8   good
chromium@399951                3.4125   0.0744024  8   good
chromium@399962                3.375    0.0707107  8   good
chromium@399965                3.43333  0.0707107  18  good
chromium@399966                3.43333  0.0923548  18  good
chromium@399966,v8@6073a342f9  3.4125   0.0517549  8   good
chromium@399966,v8@b6aa77d95f  3.5875   0.140789   8   bad
chromium@399967                3.50556  0.0921777  18  bad
chromium@399968                3.61     0.151658   5   bad
chromium@399973                3.525    0.0707107  8   bad

Bisect job ran on: win_x64_perf_bisect
Bug ID: 621173

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests memory.long_running_idle_gmail_background_tbmv2
Test Metric: Idle-v8-gc-incremental-finalize_pct_090/Idle-v8-gc-incremental-finalize_pct_090
Relative Change: 4.13%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1407
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004061586991020800


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5054991719464960

| 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 Tests>AutoBisect.  Thank you!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9003968202996035872

===== BISECT JOB RESULTS =====
Status: failed


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev    N   Good?
chromium@399929                3.425    0.138873   8   good
chromium@399951                3.4125   0.0744024  8   good
chromium@399962                3.375    0.0707107  8   good
chromium@399965                3.43333  0.0707107  18  good
chromium@399966                3.43333  0.0923548  18  good
chromium@399966,v8@6073a342f9  3.4125   0.0517549  8   good
chromium@399966,v8@b6aa77d95f  3.5875   0.140789   8   bad
chromium@399967                3.50556  0.0921777  18  bad
chromium@399968                3.61     0.151658   5   bad
chromium@399973                3.525    0.0707107  8   bad

Bisect job ran on: win_x64_perf_bisect
Bug ID: 621173

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests memory.long_running_idle_gmail_background_tbmv2
Test Metric: Idle-v8-gc-incremental-finalize_pct_090/Idle-v8-gc-incremental-finalize_pct_090
Relative Change: 4.13%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1407
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004061586991020800


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5054991719464960

| 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 Tests>AutoBisect.  Thank you!

Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/9003875150773528784

===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : [ic] Enable new LoadGlobalIC machinery.
Author  : ishell
Commit description:
  
BUG=chromium:576312
LOG=Y

Review-Url: https://codereview.chromium.org/2066763004
Cr-Commit-Position: refs/heads/master@{#37006}
Commit  : b6aa77d95f1da91407bc10d8d3344293cc7f3383
Date    : Wed Jun 15 13:32:53 2016


===== TESTED REVISIONS =====
Revision                       Mean     Std Dev    N   Good?
chromium@399929                3.45833  0.0996205  12  good
chromium@399951                3.43333  0.0937437  12  good
chromium@399962                3.39815  0.262195   27  good
chromium@399965                3.45741  0.114105   27  good
chromium@399966                3.47222  0.151679   18  good
chromium@399966,v8@6073a342f9  3.46111  0.127827   18  good
chromium@399966,v8@9df23510ea  3.41     0.0894427  5   good
chromium@399966,v8@b6aa77d95f  3.52778  0.0548319  18  bad    <--
chromium@399967                3.57778  0.095828   18  bad
chromium@399968                3.52778  0.152753   27  bad
chromium@399973                3.57778  0.14061    18  bad

Bisect job ran on: win_x64_perf_bisect
Bug ID: 621173

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --also-run-disabled-tests memory.long_running_idle_gmail_background_tbmv2
Test Metric: Idle-v8-gc-incremental-finalize_pct_090/Idle-v8-gc-incremental-finalize_pct_090
Relative Change: 3.48%
Score: 90.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1415
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9004061586991020800


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5054991719464960

| 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 Tests>AutoBisect.  Thank you!

Perf sheriff ping

ishell@, does it it seem like that CL ([ic] Enable new LoadGlobalIC machinery) could have affected memory usage? If so, was it expected?

The memory regression is not actually expected. I'll look at it.

Note that the regression is in GC time, not memory.

The Idle-v8-gc-incremental-finalize_pct_090/Idle-v8-gc-incremental-finalize_pct_090 was renamed to v8-gc-incremental-finalize_pct_090.

I would close this issue as won't fix since the regression is small (0.1ms).

Ah. Thanks. WontFix as per #23.