New issue
Advanced search Search tips

Issue 730209 link

Starred by 1 user

Issue metadata

Status: Duplicate
Merged: issue 727303
Owner: ----
Closed: Jun 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

79.2% regression in media.tough_video_cases at 475126:475226

Project Member Reported by chcunningham@chromium.org, Jun 6 2017

Issue description

This is actually a return to previous values from ~1k revisions back. Likely noise or some improvement was reverted. 
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=730209

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg5tjAqAoM


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

chromium-rel-mac12
Mergedinto: 727303
Status: Duplicate (was: Untriaged)

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

Suspected Commit
  Author : Brian Salomon
  Commit : 1c214313248a4b5a69af14539608c54fb67c2bf8
  Date   : Thu May 25 20:03:07 2017
  Subject: Revert "Clear atlas textures at creation in Chrome"

Bisect Details
  Configuration: mac_10_12_perf_bisect
  Benchmark    : media.tough_video_cases
  Metric       : vm_resident_set_size_delta_size/tulip2.mp4_gpu
  Change       : 58.60% | 6047.33333333 -> 9591.33333333

Revision                             Result                  N
chromium@475125                      6047.33 +- 2469.03      6      good
chromium@475176                      5544.0 +- 2882.14       6      good
chromium@475201                      5908.0 +- 2202.78       6      good
chromium@475214                      5875.33 +- 2138.86      6      good
chromium@475217                      5638.0 +- 1881.2        6      good
chromium@475218                      6773.33 +- 1302.29      6      good
chromium@475218,skia@9b51d1484a      5244.67 +- 1737.72      6      good
chromium@475218,skia@33fa454b16      6030.0 +- 1582.19       6      good
chromium@475218,skia@1c21431324      9491.33 +- 2334.78      6      bad       <--
chromium@475218,skia@fabe0b26d0      10237.3 +- 2476.54      6      bad
chromium@475218,skia@e7ba8b05d2      9974.67 +- 2814.37      6      bad
chromium@475218,skia@c020f17127      10516.0 +- 2773.2       6      bad
chromium@475219                      10405.3 +- 2245.81      6      bad
chromium@475220                      9858.0 +- 2677.53       6      bad
chromium@475226                      9591.33 +- 1254.51      6      bad

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

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8977499200078281280

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5573069919223808


| 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