New issue
Advanced search Search tips

Issue 762522 link

Starred by 1 user
Status: Duplicate
Merged: issue 759436
Owner: ----
Closed: Sep 2017
Cc:
sunn...@chromium.org
smcgruer@chromium.org
ericrk@chromium.org
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 1
Type: Bug



Sign in to add a comment

Codesearch on large monitor showing tiles from other layers

Project Member Reported by flackr@chromium.org, Sep 6 2017 
Chrome Version: 62.0.3198.0 (Official Build) dev (64-bit)
OS: Linux

What steps will reproduce the problem?
(1) Load codesearch page (e.g. https://cs.chromium.org/chromium/src/cc/scheduler/compositor_timing_history.cc?type=cs&q=drawinterval&sq=package:chromium&l=826 )
(2) Ensure window is wide (in my case 2560)

What is the expected result?
No random tiles or black.

What happens instead?
See black tiles and random tiles from other layers on the page.

Looks like software rasterization from chrome://gpu
Rasterization: Software only. Hardware acceleration disabled

Please use labels and text to provide additional information.

For graphics-related bugs, please copy/paste the contents of the about:gpu
page at the end of this report.
codesearch-tiles.png
417 KB View Download

Comment 1 by smcgruer@chromium.org, Sep 6 2017 

I can also reproduce this on 62.0.3198.0 (Official Build) dev (64-bit), but *not* on 61.0.3163.79 (Official Build) beta (64-bit) or 60.0.3112.113 (Official Build) (64-bit).

In my case, it appears to break as I expand the window over the boundary between my two monitors. The broken rendering however appears on both sides of the monitor-boundary, so it is not exactly at the boundary that it becomes broken. Interestingly I can then 'maximize' the Chrome window back to a single monitor and it renders correctly, and 'restore' it back to multiple windows and the rendering breaks again.

chrome://gpu output attached.
Screenshot from 2017-09-06 09:58:23.png
271 KB View Download
chrome_gpu.html
61.9 KB View Download

Comment 2 by enne@chromium.org, Sep 6 2017

Cc: sunn...@chromium.org ericrk@chromium.org
Thanks! I can repro this as well on 62.0.3198.0 official but was unable to on a personal r500004 (ee5383ccf46a8f21885752d11cf192c6740cf0b7) debug build.

Comment 3 by enne@chromium.org, Sep 6 2017

Mergedinto: 759436
Status: Duplicate (was: Untriaged)
Bisected to 498303..498313, so extremely likely to be  issue 759436 .
►

Sign in to add a comment