System Information
Hardware: Lenovo ThinkPad E14 Gen 6 (21M70015CK)
BIOS: R2JET44W (1.21), 2025-10-28
CPU: Intel Core Ultra 7 155H (Meteor Lake)
GPU: Intel Meteor Lake-P [Intel Arc Graphics] [8086:7d55] (rev 08), no
dedicated GPU
RAM: 32 GB
OS: Ubuntu 24.04.4 LTS (noble)
Kernel: 6.17.0-14-generic #14~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC
Display server: Wayland (GNOME Shell)
Mesa: 25.2.8-0ubuntu0.24.04.1
Chrome: 145.0.7632.116
Monitors: 2x external 1920x1080 (DP-6, DP-7), laptop display disabled
Firmware Versions
GuC: i915/mtl_guc_70.bin version 70.36.0 (xe driver recommends 70.44.1)
HuC: i915/mtl_huc_gsc.bin version 8.5.4
linux-firmware: 20240318.git3b128b60-0ubuntu2.25
Problem Description
The system freezes randomly during normal desktop usage, typically within 1-8
hours. The UI becomes completely unresponsive — hard reboot is required.
The problem is strongly correlated with GPU hardware acceleration in
Chromium-based applications (Google Chrome, Electron apps like Claude Desktop).
Disabling hardware acceleration in Chrome ("Use graphics acceleration when
available" = off) makes the system stable, but this is not a viable workaround
for my use case as a developer.
The problem is more frequent (and possibly only occurs) when using multiple
monitors. My typical setup is 2 external displays at 1920x1080 connected via
USB-C/DisplayPort.
I have been experiencing this issue for approximately 1 years, since
installing Ubuntu 24.04 on this machine. I am not certain which kernel
version the problem first appeared on, as I did not track it at the time.
The issue has persisted across all kernel upgrades up to the current 6.17.
Reproduction
Boot Ubuntu 24.04 on Lenovo ThinkPad E14 Gen 6 (Intel 155H) with kernel 6.17
Connect 2 external monitors
Open Google Chrome with hardware acceleration enabled
Use Chrome normally (multiple tabs, web apps)
Within 1-8 hours the system freezes
Error Logs
With i915 driver (default):
kernel: i915 0000:00:02.0: [drm] *ERROR* GT0: GUC: TLB invalidation response
timed out for seqno 62
kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 12:0:00000000
kernel: i915 0000:00:02.0: [drm] GT0: Resetting chip for stopped heartbeat on
rcs0
kernel: Fence expiration time out i915-0000:00:02.0:chrome[5652]:3ffd4!
kernel: Fence expiration time out i915-0000:00:02.0:Xwayland[5384]:4a3e!
kernel: Fence expiration time out i915-0000:00:02.0:gnome-shell[3930]:801bc!
kernel: i915 0000:00:02.0: [drm] GT0: GuC firmware i915/mtl_guc_70.bin version
70.36.0
kernel: i915 0000:00:02.0: [drm] GPU HANG: ecode 12:1:85dffffb, in Xwayland
[5384]
kernel: i915 0000:00:02.0: [drm] Xwayland[5384] context reset due to GPU hang
With xe driver (i915.force_probe=!7d55 xe.force_probe=7d55):
kernel: xe 0000:00:02.0: [drm] *ERROR* GT0: TLB invalidation fence timeout,
seqno=96187 recv=96186
kernel: xe 0000:00:02.0: [drm] GT0: trying reset from
guc_exec_queue_timedout_job [xe]
kernel: xe 0000:00:02.0: [drm] GT0: reset started
kernel: xe 0000:00:02.0: [drm] GT0: reset done
kernel: xe 0000:00:02.0: [drm] GT0: VM job timed out on non-killed execqueue
kernel: WARNING: CPU: 10 PID: 27821 at drivers/gpu/drm/xe/xe_guc_submit.c:1293
guc_exec_queue_timedout_job+0x373/0xbf0 [xe]
kernel: xe 0000:00:02.0: [drm] GT0: Timedout job: seqno=56036, lrc_seqno=56036,
guc_id=7, flags=0x0 in chrome [6614]
What I Have Tried (none of these resolved the issue)
Switching to xe driver (i915.force_probe=!7d55 xe.force_probe=7d55) — same TLB
invalidation timeout, just through xe instead of i915
Kernel parameters: i915.enable_psr=0, i915.enable_dc=0, i915.enable_fbc=1,
iommu.strict=1 — no effect
IRQ affinity + C-state pinning (pinning GPU IRQ to a dedicated core with deep
C-states disabled) — no effect on GPU hangs
Various i915.enable_guc values (0, 1, 2, 3) — either no effect or system fails
to boot
Key Observations
The TLB invalidation timeout occurs with both i915 and xe drivers, suggesting
the issue is in the shared GuC firmware or hardware level
The installed GuC firmware (70.36.0) is older than the xe driver's recommended
version (70.44.1), but Ubuntu 24.04's linux-firmware package
(20240318.git3b128b60-0ubuntu2.25) does not include the newer version
Chrome/Chromium GPU process is consistently referenced in the hang messages
The problem is confirmed as a regression — kernel 6.8 does not exhibit this
issue (as reported by other users in this bug)
Multi-monitor setup appears to increase the likelihood of the hang
Kernel Command Line
BOOT_IMAGE=/boot/vmlinuz-6.17.0-14-generic
root=UUID=0824e55d-bbf3-4f02-ba85-1081385e2b06 ro quiet splash
i915.force_probe=!7d55 xe.force_probe=7d55 vt.handoff=7
lspci
00:02.0 VGA compatible controller [0300]: Intel Corporation Meteor Lake-P
[Intel Arc Graphics] [8086:7d55] (rev 08)
Subsystem: Lenovo Meteor Lake-P [Intel Arc Graphics] [17aa:50e1]
Kernel driver in use: xe
Kernel modules: i915, xe
** Attachment added: "kernel-log-crash.txt"
https://bugs.launchpad.net/ubuntu/+source/linux-signed-oem-6.8/+bug/2085434/+attachment/5948686/+files/kernel-log-crash.txt
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2085434
Title:
i915 GPU hangs on Intel Meteor Lake [*ERROR* GT0: GUC: TLB
invalidation response timed out for seqno ...]
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2085434/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
