[mako] screen corruption/tearing after using the device for medium durations

Bug #1383745 reported by Andrew Hayzen
14
This bug affects 3 people
Affects Status Importance Assigned to Milestone
Mir
New
Medium
Unassigned
mir (Ubuntu)
New
Medium
Unassigned

Bug Description

I had the device on for a day or so and it was 'sleeping' for long periods during this duration.

What happens is the screen is visibly corrupting as things change on the display, eg scrolling up/down the dash or moving the spread. I have taken a short video [0] where you can see the corruption when moving the spread.

I have seen this on both utopic-proposed and rtm-proposed ~3 times in the last few weeks -> last month.

$ system-image-cli -i
current build number: 99
device name: mako
channel: ubuntu-touch/ubuntu-rtm/devel-proposed
alias: ubuntu-touch/ubuntu-rtm/14.09-proposed
last update: 2014-10-19 07:24:09
version version: 99
version ubuntu: 20141019
version device: 20141017.2
version custom: mako-1.1

0 - https://drive.google.com/file/d/0B3XynHVKfrvMNm00QllIWG9DY28

Tags: mako nexus4
tags: added: mako
Changed in mir:
importance: Undecided → High
Revision history for this message
Daniel van Vugt (vanvugt) wrote :

Medium at most I think.

It's a very rare bug. Most mako owners never see it. And even when you do, no functionality is lost.

Changed in mir:
importance: High → Medium
Revision history for this message
Andrew Hayzen (ahayzen) wrote :

FYI we've seen this on two further separate mako devices this morning, one of them was on #102 rtm-proposed and had an uptime 2 days+.

Revision history for this message
Michael Hall (mhall119) wrote :

I've had this all the way up to RTM #12. Just installed RTM #13 and will report if it's still there

Revision history for this message
kevin gunn (kgunn72) wrote :

follow on from an irc chat, mhall seeing this recently with uReadIt app (his most used)
if anyone sees it, it'd be useful to know if a lightdm restart corrects the problem or not.

Revision history for this message
Daniel van Vugt (vanvugt) wrote :

It does seem likely we're probably doing something wrong in our Android fencing code but I'm slightly suspicious of the mako hardware too.

See also; another mako corruption issue I found recently: bug 1406725. It's a different bug because it's a different kind of corruption occurring under different circumstances to this.

Revision history for this message
Michael Hall (mhall119) wrote :

I had the graphics issue again and restarting lightdm fixed it

http://paste.ubuntu.com/9752519/ shows RAM usage before and after

Revision history for this message
Andrew Hayzen (ahayzen) wrote :

FYI, I had this again on the latest rtm-proposed image and I can also confirm that restarting lightdm fixed the issue.

I see a similar drop in memory usage before and after [0], but this could just be due to apps closing (I had gmail and telegram open before restarting lightdm)

0 - http://pastebin.ubuntu.com/10673351/

$ system-image-cli -i
current build number: 214
device name: mako
channel: ubuntu-touch/ubuntu-rtm/devel-proposed
alias: ubuntu-touch/ubuntu-rtm/14.09-proposed
last update: 2015-03-23 01:43:53
version version: 214
version ubuntu: 20150312
version device: 20150116
version custom: mako-1.1

tags: added: nexus4
Revision history for this message
Michał Sawicz (saviq) wrote :

Syncing task from Mir.

Changed in mir (Ubuntu):
importance: Undecided → Medium
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.