Thank you, Annie!

Note that the devirtualization went into effect on Linux, not on Mac, as it was stated in the list of labels for this issue.

=== Auto-CCing suspected CL author krasin@google.com ===

Hi krasin@google.com, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


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


===== SUSPECTED CL(s) =====
Subject : Roll clang 269902:270823.
Author  : krasin
Commit description:
  
Ran `./tools/clang/scripts/upload_revision.py 270823`.

BUG= 614933 

Review-Url: https://codereview.chromium.org/2015793003
Cr-Commit-Position: refs/heads/master@{#396388}
Commit  : 709225ae8e282d53833e01e52d552d1178917951
Date    : Fri May 27 04:41:48 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@396376  3289.34  34.3785  5  good
chromium@396383  3372.05  37.1767  5  good
chromium@396386  3356.27  66.1748  5  good
chromium@396387  3353.41  32.6391  5  good
chromium@396388  3576.94  23.1922  5  bad    <--
chromium@396389  3547.28  29.6209  5  bad
chromium@396401  3519.14  36.4953  5  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 617283

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.layout
Test Metric: flexbox-column-nowrap/flexbox-column-nowrap
Relative Change: 6.99%
Score: 99.5

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6517
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9010842345418190048


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5875251582337024

| 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!

Note the bisect text is a little confusing when bisecting for improvements, but it looks like your CL is indeed the cause.

Thank you for this note, I was quite puzzled to see that devirtualization slowed down something. :)

Shall we expect more updates like this from the bisect job, or that's the final list of 1?

I kicked off 3 or 4 bisects I think, so we'll get a few updates.

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


===== SUSPECTED CL(s) =====
Subject : Roll clang 269902:270823.
Author  : krasin
Commit description:
  
Ran `./tools/clang/scripts/upload_revision.py 270823`.

BUG= 614933 

Review-Url: https://codereview.chromium.org/2015793003
Cr-Commit-Position: refs/heads/master@{#396388}
Commit  : 709225ae8e282d53833e01e52d552d1178917951
Date    : Fri May 27 04:41:48 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N  Good?
chromium@396376  7520.7   38.1124  5  good
chromium@396383  7521.04  46.8872  5  good
chromium@396386  7499.28  47.0148  5  good
chromium@396387  7553.46  30.9372  5  good
chromium@396388  8166.41  59.7394  5  bad    <--
chromium@396389  8194.06  70.6947  5  bad
chromium@396401  8107.85  62.5183  5  bad

Bisect job ran on: linux_perf_bisect
Bug ID: 617283

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests blink_perf.layout
Test Metric: multicol_tall-content-short-columns/multicol_tall-content-short-columns
Relative Change: 7.81%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/linux_perf_bisect/builds/6518
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9010842330364838592


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5773297380229120

| 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!

Annie,

so we got 2 confirmations. What's the fate of 1-2 more bisect jobs? Did they silently die?

Looks like I only ran 3 bisect jobs:
https://chromeperf.appspot.com/group_report?bug_id=617283

The third did not finish, it but it narrowed down the range to no improvement at 396386 and improvement at 396389. I think we can close the bug, since 3 bisects presented zero evidence that this is any other CL.

Sounds good. Thank you for triggering those jobs.