*** This bug is a duplicate of bug 2148538 ***
https://bugs.launchpad.net/bugs/2148538
I'm experiencing what appears to be the same underlying issue on
different hardware (not a Framework Laptop), which suggests this is not
device-specific but affects Phoenix/Rembrandt APUs more broadly.
## Hardware
- **Laptop:** ASUSTeK Zenbook 15 UM3504DA
- **APU (iGPU):** AMD Ryzen 7 7735U, Radeon 680M (Rembrandt, PCI 0000:73:00.0,
device 0x1681, DCN 3.1.2, gfx_v10_0)
- **RAM:** 16 GiB, VRAM 512M + 7355M GTT
- **Display setup:** Internal eDP panel only, PSR supported (DC PSR ver 0, sink
PSR ver 1)
## Software
- **OS:** Ubuntu 26.04.1 LTS
- **Kernel:** 7.0.0-30-generic
- **Mesa:** 26.0.8-1ubuntu0.3
- **Desktop:** GNOME 50 on Wayland
## Crash pattern
The crash follows the same multi-stage pattern described in this bug,
triggered by GPU-accelerated userspace processes (observed with both
Chrome and an Electron-based terminal app):
### Stage 1: gfxhub page fault
```
amdgpu 0000:73:00.0: [gfxhub] page fault (src_id:0 ring:24 vmid:5 pasid:57)
amdgpu 0000:73:00.0: Process tabby pid 9436 thread tabby:cs0 pid 9448
amdgpu 0000:73:00.0: in page starting at address 0x0000000000000000 from
client 0x1b (UTCL2)
amdgpu 0000:73:00.0: GCVM_L2_PROTECTION_FAULT_STATUS:0x00501431
amdgpu 0000:73:00.0: Faulty UTCL2 client ID: SQC (data) (0xa)
amdgpu 0000:73:00.0: MORE_FAULTS: 0x1
amdgpu 0000:73:00.0: PERMISSION_FAULTS: 0x3
```
### Stage 2: second ring times out (this time in gnome-shell, not the
triggering app)
```
amdgpu 0000:73:00.0: ring gfx_0.1.0 timeout, signaled seq=402345, emitted
seq=402347
amdgpu 0000:73:00.0: Process gnome-shell pid 5563 thread gnome-shel:cs0 pid
5625
amdgpu 0000:73:00.0: Starting gfx_0.1.0 ring reset
amdgpu 0000:73:00.0: Ring gfx_0.1.0 reset failed
amdgpu 0000:73:00.0: [drm] *ERROR* Failed to initialize parser -125!
```
### Stage 3: full session crash
Within ~10 seconds of the failed ring reset, gdm/gnome-session restart and I'm
dropped back to the login screen with all applications closed. No hard reboot
is required in my case (unlike the Framework 16 report, where the display goes
permanently black) — the session simply dies and gdm comes back up.
## Notes
- I've reproduced this with **multiple different GPU-accelerated applications**
(Chrome, an Electron-based terminal, PhpStorm), which matches this bug's
premise that it's not tied to a specific app but to sustained/repeated GPU
workload.
- I tried disabling PSR via `amdgpu.dcdebugmask=0x10` as a workaround — **this
did not prevent the crash**, consistent with this report's note that
`amdgpu.dcdebugmask=0x410` didn't help either.
- Occurrences are intermittent — sometimes the ring reset succeeds and only the
triggering app's GPU context is lost (recoverable), other times (as in the log
above) it escalates to a full session crash.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2147367
Title:
amdgpu: MES firmware intermittently unresponsive on Phoenix APU
(gfx1103), leading to unrecoverable GPU hang on Framework 16
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-oem-6.17/+bug/2147367/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
