Bug 101206 - intermittent lock up after login to GNOME under wayland
Summary: intermittent lock up after login to GNOME under wayland
Status: NEW
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: unspecified
Hardware: Other All
: medium normal
Assignee: Nouveau Project
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-05-26 19:59 UTC by Adi
Modified: 2017-05-27 21:31 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
last 260 lines of journalctl log 2017-05-26 (31.40 KB, text/plain)
2017-05-26 19:59 UTC, Adi
no flags Details
last 275 lines of journalctl log 2017-05-11 (31.23 KB, text/plain)
2017-05-26 20:00 UTC, Adi
no flags Details
last 260 lines of journalctl log 2017-05-11 (41.81 KB, text/plain)
2017-05-26 20:09 UTC, Adi
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Adi 2017-05-26 19:59:35 UTC
Created attachment 131529 [details]
last 260 lines of journalctl log 2017-05-26

my notebook locks up intermittently directly after login in to GNOME running under wayland.

I'm running Fedora 24 with all updates installed. Kernel is 4.10.14, although I've had this problem at least since 4.9, if not before.

Attached are two logs from journalctl from when the locks up happens. I've only attached just enough so I don't have to zip the logs. If needed I can attached complete logs as zips or compressed tars.

The notebook also sometimes locks up (although not as frequently) when unblanking the screen.
Comment 1 Adi 2017-05-26 20:00:28 UTC
Created attachment 131530 [details]
last 275 lines of journalctl log 2017-05-11
Comment 2 Adi 2017-05-26 20:03:02 UTC
looks like the log from 2017-05-11 doesn't contain any reference to the error. so please ignore.
Comment 3 Ilia Mirkin 2017-05-26 20:04:43 UTC
I'm guessing you have a tesla-era GPU (G80-GT21x)? If so, there's an undiagnosed issue where these sometimes "mess up". This is seen through an initial bunch of CACHE_ERROR's when trying to bring up a new channel (you can see where it's setting methods 180, etc, which are the dma objects being bound), followed by a TLB update hang.

No clue what causes it, no clue how to fix it.
Comment 4 Adi 2017-05-26 20:09:02 UTC
Created attachment 131531 [details]
last 260 lines of journalctl log 2017-05-11

here is the correct log from 2017-05-11.
This lock up can't have happend after login in to GNOME since the lock up happens at 6:40 on the 11th. The journalctl log starts on the 10th though.
Comment 5 Adi 2017-05-27 21:31:20 UTC
I also forgot to mention that it's a NVIDIA Corporation GT218M [NVS 3100M] (rev a2) chip.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.