New issue
Advanced search Search tips

Issue 788787 link

Starred by 1 user
Status: WontFix
Owner:
gurcheta...@chromium.org
Closed: Nov 2017
Cc:
ihf@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: ----
Type: Bug-Regression



Sign in to add a comment

29.8% regression in graphics_GLMark2 at 32460001006100000:32570001009800000

Project Member Reported by gurchetansingh@google.com, Nov 27 2017 
ChromeOS Version range: 64.10061.0.0 - 64.10098.0.0 
Chrome Version range: 64.0.3246.0 - 64.0.3257.0 

https://crosland.corp.google.com/log/10061.0.0..10098.0.0
https://chromium.googlesource.com/chromium/src/+log/64.0.3246.0..64.0.3257.0?pretty=fuller&n=10000

Also looks related to the Mesa uprev that happen in 10077.0.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Nov 27 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=788787

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=e9cfb39bffe5d601bb2b7b9de4ad0c7bc3f03eaf2474d2b0e657cca76e48cf86


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

cros-wizpig

Comment 2 by gurcheta...@chromium.org, Nov 27 2017

Labels: Gfx-Guard

Comment 3 by pwang@chromium.org, Nov 27 2017 

FYI, I915_EXEC_NO_RELOC and I915_EXEC_BATCH_FIRST flag is used in this mesa uprev.

Comment 4 by gurcheta...@chromium.org, Nov 30 2017 

As per ihf@, we don't care about graphics_GLMark2 micro-benchmark regressions (though we should worry if the aggregate score regresses).  graphics_GLMark2 mostly checks for correctness, not performance.

Comment 5 by gurcheta...@chromium.org, Nov 30 2017

Status: WontFix (was: Assigned)
►

Sign in to add a comment