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

Issue 827139 link

Starred by 2 users
Status: Duplicate
Merged: issue 827144
Owner:
khushals...@chromium.org
Closed: Apr 2018
Cc:
█ seobrien@google.com
█ fsam...@chromium.org
enne@chromium.org
schenney@chromium.org
█ kraynov@chromium.org
khushalsagar@google.com
spang@chromium.org
chrishtr@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

2.3% regression in system_health.memory_mobile at 546255:546359

Project Member Reported by kraynov@chromium.org, Mar 29 2018 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Mar 29 2018 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=827139

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=6c565e8388ce2be371cadd44b4265ca9fc46ebdf6d5d510f28e2c1a119f43117


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

android-nexus5X
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Mar 29 2018

Cc: seobrien@google.com chrishtr@chromium.org khushalsagar@google.com fsam...@chromium.org schenney@chromium.org spang@chromium.org enne@chromium.org
Status: Assigned (was: Untriaged)
📍 Found significant differences after each of 3 commits.
https://pinpoint-dot-chromeperf.appspot.com/job/14d5b0b7440000

[PE] Re-compute compositing reasons if used transform-3d state changed. by chrishtr@chromium.org
https://chromium.googlesource.com/chromium/src/+/fa8585ccd6f3c904fcb0afa76ea1237d349d77e5

viz: Ensure DirectLayerTreeFrameSink frame acks are async. by khushalsagar@google.com
https://chromium.googlesource.com/chromium/src/+/4361dcb9a14ccbf76b21abc3c3592a905430c3b1

Reland "libgestures_glue: Send MSC_TIMESTAMP if available" by seobrien@google.com
https://chromium.googlesource.com/chromium/src/+/3ad53273b8dae78596f976e92c4fac0722842a05

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Comment 4 by sullivan@chromium.org, Apr 5 2018

Owner: khushals...@chromium.org
There were multiple changes in this range:

khushalsagar: Assigning to you since there is a ~1MB regression reproduced at "viz: Ensure DirectLayerTreeFrameSink frame acks are async.".

"Reland "libgestures_glue: Send MSC_TIMESTAMP if available" regressed about 400KiB.


"[PE] Re-compute compositing reasons if used transform-3d state changed." improved memory usage.
Project Member

Comment 6 by 42576172...@developer.gserviceaccount.com, Apr 5 2018

Cc: bauerb@chromium.org tmartino@chromium.org kouhei@chromium.org horo@chromium.org
Owner: horo@chromium.org
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/14c6db04c40000

Show a bad feature flag infobar only for flags in about:flags on Android by horo@chromium.org
https://chromium.googlesource.com/chromium/src/+/5555f7d5e3b1efeffc735ac2dbbfe286faaf0b58

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions

Comment 7 by khushals...@chromium.org, Apr 5 2018

Cc: -kouhei@chromium.org -tmartino@chromium.org -bauerb@chromium.org -horo@chromium.org
Mergedinto: 827144
Owner: khushals...@chromium.org
Status: Duplicate (was: Assigned)
I had started the bisect above to see if a similar recovery after this regression was from a related change. It isn't, and this looks the same root cause as  issue 827144 .
►

Sign in to add a comment