New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 681101 link

Starred by 1 user

Issue metadata

Status: Fixed
Merged: issue 680776
Owner:
Last visit > 30 days ago
Closed: Jan 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

491.5% regression in blink_perf.canvas at 442934:443046

Project Member Reported by nzolghadr@chromium.org, Jan 13 2017

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=681101

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDgoPyvrgoM


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

chromium-rel-win7-gpu-intel
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Jan 13 2017

Mergedinto: 680776
Status: Duplicate (was: Untriaged)

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

Suspected Commit
  Author : zakerinasab
  Commit : 7fb9535883e50ab0d7262ea8a0e8470934e57a9c
  Date   : Wed Jan 11 19:09:13 2017
  Subject: Implement color management for ImageData

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : blink_perf.canvas
  Metric       : toBlob_duration_jpeg/toBlob_duration_jpeg
  Change       : 471.07% | 205.728166667 -> 1174.84966667

Revision             Result                  N
chromium@442934      205.728 +- 15.2825      6      good
chromium@442962      201.707 +- 16.1474      6      good
chromium@442969      203.41 +- 8.58663       6      good
chromium@442971      200.817 +- 14.5922      6      good
chromium@442972      204.118 +- 24.3293      6      good
chromium@442973      1299.7 +- 259.934       6      bad       <--
chromium@442976      1305.68 +- 65.7136      6      bad
chromium@442990      1258.38 +- 242.928      6      bad
chromium@443046      1174.85 +- 79.7868      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 blink_perf.canvas

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8990556515980639120

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5239004733112320


| 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!
Owner: zakerinasab@chromium.org
Status: Assigned (was: Duplicate)
Un-duping another performance regression to investigate when relanding this CL.
Cc: sullivan@chromium.org
Status: Fixed (was: Assigned)
The issue has been addressed and the CL is landed again at commit position #443939.(https://codereview.chromium.org/2555213002). Here is the link to the Chrome perf graph showing that the regression is fixed:

https://chromeperf.appspot.com/report?sid=cfbb0afcf875859b1f18dda277b66fff2dfe850922e598bd84d0b3e760811c55&start_rev=442200&end_rev=444200

Sign in to add a comment