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

Issue 923199 link

Starred by 1 user
Status: WontFix
Owner:
kylec...@chromium.org
Closed: Jan 18
Cc:
cros-perf-detectives@google.com
khmel@chromium.org
█ skuhne@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 2
Type: Bug-Regression



Sign in to add a comment

cheets_GamePerformance/opengl_fps regression on minnie since R72-11289.0.0

Project Member Reported by kcwu@chromium.org, Jan 18 (5 days ago) 
cheets_GamePerformance/opengl_fps
regressed from 100 to 60 (higher is better)
https://screenshot.googleplex.com/g21VcMJkM3L.png
This is minnie only. Other boards are good.

According to our bisect result
https://crosperf.googleplex.com/b/54d13ab9-3610-421d-9595-b148df49d562
this regression is introduced by this CL
https://chromium.googlesource.com/chromium/src.git/+/10d7fc3b763a758374e0dfc74d5fd9d36bc1b146
  [Ozone DRM] Don't round refresh rate for PresentationFeedback

kylechar, could you please take a look?

Comment 1 by kylechar@google.com, Jan 18 (4 days ago)

Status: WontFix (was: Untriaged)
This is WAI basically. That CL was to fix b/116769488 where arc++ was running at ~100fps instead of ~60fps. I had changed how the refresh rate was plumbed out of the display compositor and the refresh rate was getting rounded to an integer instead of using the floating point value. This interacted badly with scheduling code and things ran much faster than they should wasting power.

Comment 2 by cywang@google.com, Jan 21 (2 days ago)

Cc: skuhne@chromium.org khmel@chromium.org
+Yury, Stefan
FYI

Comment 3 by khmel@chromium.org, Yesterday (34 hours ago) 

#2 #1 - Thank you for explanation. This sounds reasonable.
►

Sign in to add a comment