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

Issue 596035 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: May 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Android
Pri: 2
Type: Bug-Regression



Sign in to add a comment

smoothness.gpu_rasterization.top_25_smooth[.reference] failure on Android Nexus6 Perf (2) at 381864:381892

Project Member Reported by petrcermak@chromium.org, Mar 18 2016

Issue description

Revision range first seen: https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus6%20Perf%20%282%29/builds/2214
Link to failing step log: https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus6%20Perf%20%282%29/builds/2214/steps/smoothness.gpu_rasterization.top_25_smooth/logs/stdio

In all cases, the failure is due to a native crash on https://mail.google.com/mail/.

The next steps (https://chromium.googlesource.com/chromium/src/+/master/tools/perf/docs/perf_bot_sheriffing.md#testfailures):
  1. Disable the benchmark on Nexus 6.
  2. Kick off a bisect.
  3. Once the benchmark is disabled, downgrade this bug to Pri-2 and re-assign it to the test owner.
 
Project Member

Comment 2 by 42576172...@developer.gserviceaccount.com, Mar 18 2016


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


===== SUSPECTED CL(s) =====
Subject : Reland - Update DevTools Tracing.Start to accept trace config as a parameter
Author  : zhenw
Commit description:
  
Original CL: https://crrev.com/1765153002

Forgot to decouple some statement from DCHECK, which was actually
pointed out by Primiano before. Should be fixed now.

--------------------
original description
--------------------

This CL updates Tracing.Start to accept trace config as
a parameter when starting tracing. It is backward compatible
with the old way.

Design doc: https://goo.gl/GxQ23k
BUG= 579358 
TBR=primiano@chromium.org,caseq@chromium.org,simonhatch@chromium.org,petrcermak@chromium.org,pfeldman@chromium.org

Review URL: https://codereview.chromium.org/1808353002

Cr-Commit-Position: refs/heads/master@{#381892}
Commit  : af6fe5172a632660bb0c2d70b6b7f3bbe08aa284
Date    : Fri Mar 18 05:00:14 2016


===== TESTED REVISIONS =====
Revision                Exit Code   Std. Dev.   Num Values  Good?
chromium@381863         0           N/A         5           good
chromium@381878         0           N/A         5           good
chromium@381885         0           N/A         5           good
chromium@381889         0           N/A         5           good
chromium@381891         0           N/A         5           good
chromium@381892         0           N/A         5           bad

Bisect job ran on: android_nexus6_perf_bisect
Bug ID: 596035

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --also-run-disabled-tests smoothness.gpu_rasterization.top_25_smooth
Test Metric: mean_frame_time/mean_frame_time
Relative Change: 0.00%
Score: 0.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2026
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9017851858650952480


| 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 label Cr-Tests-AutoBisect.  Thank you!
The bisect failed to find any cuplrit (all exit codes were 0), so I'm running another one: https://chromeperf.appspot.com/buildbucket_job_status/9017839628230091184

In the meantime, the bot has gone green again (https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus6%20Perf%20%282%29/builds/2216), so I unticked the disabling patch (https://codereview.chromium.org/1813913003/). If it stays green, there's no point in disabling the benchmark.

So far, I haven't been able to figure out why the benchmark has gone either red or green.
Sorry for the confusion, Petr! I filed  bug 596086  about the confusing output in #2.
Cc: -vmi...@chromium.org petrcermak@chromium.org
Labels: -Pri-1 Pri-2
Owner: vmi...@chromium.org
The benchmark was green twice in a row now, so it looks like the issue is gone. Therefore, I'm lowering the priority of this bug and assigning it to the test owner to decide if anything needs to be done.
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Mar 18 2016


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


===== SUSPECTED CL(s) =====
Subject : Reland - Update DevTools Tracing.Start to accept trace config as a parameter
Author  : zhenw
Commit description:
  
Original CL: https://crrev.com/1765153002

Forgot to decouple some statement from DCHECK, which was actually
pointed out by Primiano before. Should be fixed now.

--------------------
original description
--------------------

This CL updates Tracing.Start to accept trace config as
a parameter when starting tracing. It is backward compatible
with the old way.

Design doc: https://goo.gl/GxQ23k
BUG= 579358 
TBR=primiano@chromium.org,caseq@chromium.org,simonhatch@chromium.org,petrcermak@chromium.org,pfeldman@chromium.org

Review URL: https://codereview.chromium.org/1808353002

Cr-Commit-Position: refs/heads/master@{#381892}
Commit  : af6fe5172a632660bb0c2d70b6b7f3bbe08aa284
Date    : Fri Mar 18 05:00:14 2016


===== TESTED REVISIONS =====
Revision                Exit Code   Std. Dev.   Num Values  Good?
chromium@381863         0           N/A         5           good
chromium@381878         0           N/A         5           good
chromium@381885         0           N/A         5           good
chromium@381889         0           N/A         5           good
chromium@381891         0           N/A         5           good
chromium@381892         0           N/A         5           bad

Bisect job ran on: android_nexus6_perf_bisect
Bug ID: 596035

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --also-run-disabled-tests smoothness.gpu_rasterization.top_25_smooth
Test Metric: mean_frame_time/mean_frame_time
Relative Change: 0.00%
Score: 0.0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2027
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9017839628230091184


| 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 label Cr-Tests-AutoBisect.  Thank you!
Status: Fixed (was: Started)
As noted in #5, it looks like the issue is gone now.

Sign in to add a comment