New issue
Advanced search Search tips

Issue 823315 link

Starred by 4 users

Issue metadata

Status: Verified
Owner:
Closed: Mar 2018
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocking:
issue 818627



Sign in to add a comment

thread_times.key_silk_cases broken between 543349:543366

Project Member Reported by ajuma@chromium.org, Mar 19 2018

Issue description

thread_other_cpu_time_per_frame and other subtests suspiciously dropped to 0. Something's broken, needs a bisect.
 
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Mar 19 2018

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=823315

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=7aa3383a5a959f4cbaaeeb24fe171cb8133057246e8f989345f5e3e01601e211


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

android-nexus6

Comment 3 by ajuma@chromium.org, Mar 19 2018

Blocking: 818627
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, Mar 19 2018

Cc: haraken@chromium.org kinuko@chromium.org altimin@chromium.org
Owner: altimin@chromium.org
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/12b05db9440000

[scheduler] Use WorkerSchedulerProxy in DedicatedWorkerThread. by altimin@chromium.org
https://chromium.googlesource.com/chromium/src/+/053b619cf7ac1dd0ed6504e6e65da0e45d0a9d4e

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Status: Started (was: Assigned)
That's rather strange. Looking.
Cc: tdres...@chromium.org vmi...@chromium.org
 Issue 822813  has been merged into this issue.
Cc: alexilin@chromium.org
 Issue 823348  has been merged into this issue.
Project Member

Comment 8 by bugdroid1@chromium.org, Mar 19 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/826afc1b01683f467bdaf535e970c084a8d73bf3

commit 826afc1b01683f467bdaf535e970c084a8d73bf3
Author: Alexander Timin <altimin@chromium.org>
Date: Mon Mar 19 19:22:59 2018

Fix compositor thread name

Some benchmarks depend on fixed compositor thread name. Revert the name
back to the old one for compatibility.

TBR=haraken@chromium.org
BUG= 823315 

Change-Id: I07c324016280c33b240fe5f32e5c560ddd9250db
Reviewed-on: https://chromium-review.googlesource.com/969042
Reviewed-by: Alexander Timin <altimin@chromium.org>
Reviewed-by: Hector Dearman <hjd@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Hector Dearman <hjd@chromium.org>
Commit-Queue: Alexander Timin <altimin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#544110}
[modify] https://crrev.com/826afc1b01683f467bdaf535e970c084a8d73bf3/third_party/WebKit/Source/platform/WebThreadType.cpp

Cc: skyos...@chromium.org
 Issue 823352  has been merged into this issue.
Status: Fixed (was: Started)
The fix has landed, bots should start feeling better now.
Status: Verified (was: Fixed)
Can confirm that the bots feel better now, when I panicked seeing ~1500 alerts from them saying the have non-zero values now :p

Sign in to add a comment