New issue
Advanced search Search tips

Issue 714624 link

Starred by 1 user

Issue metadata

Status: WontFix
Owner:
Closed: Apr 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug-Regression



Sign in to add a comment

32.7% regression in angle_perftests at 466134:466250

Project Member Reported by jmad...@chromium.org, Apr 24 2017

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=714624

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg4pzr1wkM


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

chromium-rel-win7-gpu-nvidia
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, Apr 25 2017

Owner: jmad...@chromium.org

=== Auto-CCing suspected CL author jmadill@chromium.org ===

Hi jmadill@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


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

Suspected Commit
  Author : Jamie Madill
  Commit : 6de518589879b6b5802587bdca74bbb4f614a7f6
  Date   : Wed Apr 19 21:50:10 2017
  Subject: Optimize angle::BitSetIterator.

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : angle_perftests
  Metric       : Uniforms_gl_matrix_repeating/score
  Change       : 35.52% | 10800.5 -> 6963.83333333

Revision                              Result                  N
chromium@466133                       10800.5 +- 2677.71      12      good
chromium@466137                       10766.7 +- 2165.75      12      good
chromium@466138                       10769.3 +- 3022.67      12      good
chromium@466138,angle@a0e0aebbf2      10965.0 +- 1899.65      12      good
chromium@466138,angle@6de5185898      7124.67 +- 1492.06      12      bad       <--
chromium@466138,angle@8162926b2b      7186.5 +- 1244.06       12      bad
chromium@466139                       6943.83 +- 1938.18      12      bad
chromium@466140                       7297.33 +- 1295.43      12      bad
chromium@466148                       7251.67 +- 716.002      12      bad
chromium@466163                       6971.0 +- 1729.98       12      bad
chromium@466192                       7406.0 +- 178.65        12      bad
chromium@466250                       6963.83 +- 1816.95      12      bad

To Run This Test
  .\src\out\Release_x64\angle_perftests.exe --test-launcher-print-test-stdio=always --test-launcher-jobs=1

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981424296955664064

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4710656449708032


| 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!
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Apr 25 2017


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

Suspected Commit
  Author : Jamie Madill
  Commit : 6de518589879b6b5802587bdca74bbb4f614a7f6
  Date   : Wed Apr 19 21:50:10 2017
  Subject: Optimize angle::BitSetIterator.

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : angle_perftests
  Metric       : Uniforms_gl_matrix_repeating/score
  Change       : 33.99% | 11311.1666667 -> 7466.33333333

Revision                              Result                  N
chromium@466133                       11311.2 +- 1687.6       12      good
chromium@466137                       11022.0 +- 1898.08      12      good
chromium@466138                       11143.8 +- 1755.01      12      good
chromium@466138,angle@a0e0aebbf2      10820.8 +- 2034.26      12      good
chromium@466138,angle@6de5185898      7292.17 +- 1493.37      12      bad       <--
chromium@466138,angle@8162926b2b      7596.5 +- 1289.5        12      bad
chromium@466139                       7418.0 +- 1022.64       12      bad
chromium@466140                       7430.5 +- 1142.45       12      bad
chromium@466148                       7026.67 +- 1741.31      12      bad
chromium@466163                       7178.0 +- 1554.5        12      bad
chromium@466192                       7311.17 +- 1488.05      12      bad
chromium@466250                       7466.33 +- 1453.19      12      bad

To Run This Test
  .\src\out\Release_x64\angle_perftests.exe --test-launcher-print-test-stdio=always --test-launcher-jobs=1

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8981424303130860912

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6587192891146240


| 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!
Components: Internals>GPU>ANGLE
Status: Started (was: Untriaged)
Labels: OS-Windows
Status: WontFix (was: Started)
I can't repro this. It's kinda bad that such a large regression on the bots isn't detectable locally. Note: I am on Windows 10, with a different GPU, and different driver version.

Sign in to add a comment