New issue
Advanced search Search tips

Issue 730538 link

Starred by 1 user

Issue metadata

Status: Duplicate
Merged: issue 730119
Owner:
Closed: Jun 2017
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: ----
Type: ----



Sign in to add a comment

12.6% regression in thread_times.tough_scrolling_cases at 476916:476955

Project Member Reported by kouhei@google.com, Jun 7 2017

Issue description

This is clearly from reverting clang change.
 

Comment 1 by kouhei@google.com, Jun 7 2017

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=730538

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgpvaWsAsM


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

chromium-rel-win7-x64-dual
To clarify, this shows improvement from clang switch.

Comment 4 by h...@chromium.org, Jun 7 2017

Status: WontFix (was: Assigned)
> To clarify, this shows improvement from clang switch.

That's great! I'll mark this wontfix as there's no action to take now.

It's nice to know this will improve when we turn clang on for real.
Mergedinto: 730119
Status: Duplicate (was: WontFix)

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

Suspected Commit
  Author : hans
  Commit : ba2ff9ad0d643a443575c839e56ae9edf7106643
  Date   : Mon Jun 05 02:02:06 2017
  Subject: Revert of "win: Set is_clang=true by default over the weekend."

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : thread_times.tough_scrolling_cases
  Metric       : thread_raster_cpu_time_per_frame/thread_raster_cpu_time_per_frame
  Change       : 12.74% | 1.26563943491 -> 1.42682313542

Revision             Result                     N
chromium@476915      1.26564 +- 0.0237722       6      good
chromium@476920      1.26556 +- 0.0106612       6      good
chromium@476921      1.4375 +- 0.0146458        6      bad       <--
chromium@476922      1.42584 +- 0.010737        6      bad
chromium@476923      1.43569 +- 0.016401        6      bad
chromium@476925      1.43145 +- 0.0135174       6      bad
chromium@476935      1.43988 +- 0.0150693       6      bad
chromium@476955      1.42682 +- 0.00969412      6      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests thread_times.tough_scrolling_cases

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977450165621573328

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5784303255945216


| 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!

Sign in to add a comment