FS#71642 - Random GPU crash - linux-5.13.5/i915

Attached to Project: Arch Linux
Opened by Kurian Thayil (kmt047) - Wednesday, 28 July 2021, 19:08 GMT
Last edited by Toolybird (Toolybird) - Tuesday, 06 June 2023, 03:32 GMT
Task Type Bug Report
Category Kernel
Status Closed
Assigned To Jan Alexander Steffens (heftig)
Levente Polyak (anthraxx)
Architecture x86_64
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 1
Private No

Details

Getting random GPU crashes after an upgrade to 5.13.5-arch1-1.

[ 48520.792] (EE) intel(0): Failed to submit rendering commands (Input/output error), disabling acceleration.
[ 48520.792] (EE) intel(0): When reporting this, please include /sys/class/drm/card0/error and the full dmesg.


dmesg and /sys/class/drm/card0/error are attached.

This is on a Thinkpad T480.
This task depends upon

Closed by  Toolybird (Toolybird)
Tuesday, 06 June 2023, 03:32 GMT
Reason for closing:  Fixed
Additional comments about closing:  See comments
Comment by Bruno (pazitos10) - Friday, 30 July 2021, 23:19 GMT
I have the same kernel version and I'm getting a GPU HANG error message on my Asus S510UQ with "Intel Corporation HD Graphics 620" when trying to use Android emulator and sharing my screen through Discord. The error messages goes like this: ```i915 0000:00:02.0: [drm] GPU HANG: ecode 9:1:85dfbfff, in MainLoopThread [27078]```
The only way I've found to recover from this hang is to reboot by switching to tty mode first (please, let me know if there is another way).

In my case I have `xf86-video-intel 1:2.99.917+916+g31486f40-1`
I've attached the journal output files for my previous boot (one is complete, the other one has only the lines referring to i915 and error codes).
Comment by Neal (meltdown) - Tuesday, 03 August 2021, 23:14 GMT
I also experience graphics related crashed with kernels 5.13 and up. It's usually when I try to close Discord or Chrome if it's playing a video and the cursor hovers over the Gnome System panel in the upper-right corner. It starts by lagging and showing artifacts around the cursor, then I get a black screen due to Gnome crashing. I can hit Ctrl-C to get to GDM and log back in.
I have a Lenovo T440 with Intel Graphics 440 and an i5-4300u.

Downgrading to Linux 5.12.15 is the only way I've been able to avoid the crashes.
Comment by Neal (meltdown) - Saturday, 07 August 2021, 15:38 GMT
EDIT: Nevermind, problems started today when opening the Gnome Activities Launcher. Graphical glitches then gnome crashed.

Original msg: I've been using Linux 5.13.8 for about 24 hours and so far the issues hasn't happened. Give it a try @pazitos710 and @kmt047
Comment by Kurian Thayil (kmt047) - Wednesday, 11 August 2021, 10:51 GMT
Thank you @meltdown. Apparently it happened with me as well just now (on 5.13.8-arch1-1). So this is still an ongoing issue.

Aug 11 16:07:13 kmt-lappy kernel: i915 0000:00:02.0: [drm] Resetting bcs0 for CS error
Aug 11 16:07:13 kmt-lappy kernel: i915 0000:00:02.0: [drm] Xorg[564] context reset due to GPU hang
Aug 11 16:07:13 kmt-lappy kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:2:000e0e02, in Xorg [564]
Aug 11 16:07:25 kmt-lappy kernel: i915 0000:00:02.0: [drm] Resetting bcs0 for CS error
Aug 11 16:07:25 kmt-lappy kernel: i915 0000:00:02.0: [drm] Xorg[564] context reset due to GPU hang
Aug 11 16:07:25 kmt-lappy kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:2:0004040f, in Xorg [564]
Aug 11 16:07:47 kmt-lappy kernel: i915 0000:00:02.0: [drm] Resetting rcs0 for CS error
Aug 11 16:09:39 kmt-lappy kernel: i915 0000:00:02.0: [drm] Resetting bcs0 for CS error
Aug 11 16:09:39 kmt-lappy kernel: i915 0000:00:02.0: [drm] Xorg[564] context reset due to GPU hang
Aug 11 16:09:39 kmt-lappy kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:2:00cacac6, in Xorg [564]
Aug 11 16:10:47 kmt-lappy kernel: i915 0000:00:02.0: [drm] Resetting bcs0 for CS error
Aug 11 16:10:47 kmt-lappy kernel: i915 0000:00:02.0: [drm] Xorg[564] context reset due to GPU hang
Aug 11 16:10:47 kmt-lappy kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 9:2:fe3cfb9a, in Xorg [564
Comment by Neal (meltdown) - Saturday, 21 August 2021, 02:22 GMT
Since 5.13.10 and now 5.13.12 I've had no crashes
Comment by Kurian Thayil (kmt047) - Monday, 23 August 2021, 15:12 GMT
@meltdown, I haven't noticed either. Im still on 5.13.10 and will be upgrading tonight. Will observe for few more days.

Loading...