GPU Hang Causing GNOME Shell Freezes on Zorin OS (Intel Raptor Lake-P UHD Graphics)

Hardware Support
1

Hello everyone,

I'm experiencing intermittent UI freezes on Zorin OS and, after investigating the issue, I found that the freezes appear to be caused by GPU hangs rather than GNOME Shell itself. I'd like to share my findings in case anyone has experienced the same issue.

System Information

  • Distribution: Zorin OS (latest updates)
  • Session: Xorg
  • Kernel: 6.17.0-35-generic
  • GPU: Intel Raptor Lake-P UHD Graphics
  • Driver: i915
  • CPU: Intel 13th Gen (Raptor Lake-P)

Kernel version:

uname -r

Output:

6.17.0-35-generic

GPU:

lspci | grep VGA

Output:

00:02.0 VGA compatible controller: Intel Corporation Raptor Lake-P [UHD Graphics] (rev 04)

Current kernel parameters:

cat /proc/cmdline

Output:

BOOT_IMAGE=/boot/vmlinuz-6.17.0-35-generic root=UUID=... ro quiet splash vt.handoff=7

No custom i915 parameters were configured.

Symptoms

The system works normally most of the time, but occasionally the desktop freezes for a few seconds.

The freezes are most noticeable when:

  • Opening the Zorin application menu.
  • Interacting with the GNOME Shell interface.

The mouse pointer usually stops responding briefly, and then the desktop recovers by itself.

The system does not reboot or crash completely.

Investigation

I monitored the system logs in real time using:

journalctl -f

When the freeze occurred, I found the following messages:

i915 0000:00:02.0: [drm] *ERROR* GT0: GUC: Engine reset failed on 0:0 (rcs0)
i915 0000:00:02.0: [drm] GPU HANG: ecode 12:1:84dffffb, in gnome-shell
i915 0000:00:02.0: [drm] GT0: Resetting chip for GuC failed to reset engine mask=0x1
i915 0000:00:02.0: [drm] *ERROR* GT0: rcs0 reset request timed out
i915 0000:00:02.0: [drm] gnome-shell context reset due to GPU hang

I also checked the kernel log:

sudo dmesg | grep -Ei "guc|huc|i915"

Firmware initialization completed successfully:

GuC firmware i915/adlp_guc_70.bin version 70.36.0
HuC firmware i915/tgl_huc.bin version 7.9.3
HuC: authenticated for all workloads
GUC: submission enabled
GUC: SLPC enabled
GUC: RC enabled

However, later during normal desktop usage, multiple GPU hangs were reported:

GPU HANG: ecode 12:1:84dffffb, in gnome-shell
Engine reset failed
context reset due to GPU hang

The kernel also saved the GPU error state:

GPU error state saved to /sys/class/drm/card1/error

What I ruled out

At this point, I don't believe the issue is caused by:

  • Flatpak applications
  • Lack of RAM
  • SSD issues
  • GNOME Shell itself
  • Xorg itself

The logs consistently point to the Intel i915 graphics driver experiencing GPU hangs while GNOME Shell is rendering.

Current hypothesis

My current hypothesis is that this is related to:

  • Intel i915 driver
  • GuC submission
  • Kernel 6.17
  • Intel Raptor Lake-P integrated graphics

The freezes coincide exactly with GPU hang detection and engine reset attempts.

Next test

As a diagnostic step, I plan to boot with:

i915.enable_guc=0

to disable GuC submission and determine whether the GPU hangs disappear.

Has anyone with Intel Alder Lake or Raptor Lake graphics experienced similar GPU HANG errors on recent kernels?

Any suggestions or known fixes would be greatly appreciated.

2

Does it behave the same with Wayland? Because of this i915, maybe some Pwer Saving Feature could be an Issue. I think,. You can set a GRUB Parameter to disable Power Saving on this.

Another Option would be to try a newer Kernel. Maybe the Liquorix Kernel for Example to see if it might run better because of better Hardware Support.

3

Thank you for the suggestions.

I investigated the issue further and found that every time the UI freezes, the kernel logs report an i915 GPU hang while gnome-shell is running. The relevant messages are:

GPU HANG: ecode 12:1:84dffffb, in gnome-shell
GUC: Engine reset failed
gnome-shell context reset due to GPU hang

The GuC and HuC firmware load successfully during boot, so the issue seems to happen later during normal desktop usage.

I haven't yet tested disabling the power-saving features or the GuC submission, but that will be my next step. I'll start by testing i915.enable_guc=0, and if necessary I'll also try disabling PSR or other power-saving features.

Regarding Wayland, I have used it before, but I need to perform more controlled testing while monitoring the logs to determine whether the same GPU hang occurs there as well.

I'll report back with the results. Thanks again!

2 Likes