You are not logged in.
Pages: 1
Is anyone aware of problems with the 6.17.9 kernel? After an upgrade I started getting a lot of kernel panics. After downgrading to 6.17.8, I'm not seeing them anymore. This is not an area I'm well versed in but, I'll be happy to provide specific info. Just tell me what you need.
Last edited by MickeyRat (2025-11-29 18:07:14)
Some cause happiness wherever they go; others whenever they go.
- Oscar Wilde
Offline
Does journalctl logs from failed boots show something interesting (sudo journalctl -b-n where n is number of boots ego) ? Post full logs, if yes.
Offline
Does journalctl logs from failed boots show something interesting (sudo journalctl -b-n where n is number of boots ego) ? Post full logs, if yes.
Thanks for the reply. It didn't fail to boot. It would boot fine but, a while after I booted, it would lock up, go dark and then the LightDM login screen would come up.
That said, I misspoke in a couple ways. First it wasn't kernel panics. it was core dumps. I know BIG difference. Second at the time I originally posted, I wasn't getting core dumps with the 6.17.8 kernel but, I am now. I think these lines from the journal might indicate the problem:
Nov 29 14:40:12 groucho kernel: amdgpu 0000:0e:00.0: amdgpu: ring gfx_0.0.0 timeout, signaled seq=2002395, emitted seq=2002397
Nov 29 14:40:12 groucho kernel: amdgpu 0000:0e:00.0: amdgpu: Process endless-sky pid 6871 thread endless-sk:cs0 pid 6896
Nov 29 14:40:12 groucho kernel: amdgpu 0000:0e:00.0: amdgpu: Starting gfx_0.0.0 ring reset
Nov 29 14:40:12 groucho kernel: amdgpu 0000:0e:00.0: amdgpu: Ring gfx_0.0.0 reset failedUnless someone has a better suggestion, I suspect I'm going to have to do a reinstall but, I'll do some disk checks first.
Some cause happiness wherever they go; others whenever they go.
- Oscar Wilde
Offline
I experience a similar issue. My machine freezes with the same messages after a short amount of time. I downgraded kernel and firmware but to no avail.
Right now I have amd-ucode downgraded to 20251111 and it is running fine for 2½ hours now. I'll give it some further testing tomorrow.
You might want to try downgrading amd-ucode as well and check if the situation improves.
I opened a thread myself: https://bbs.archlinux.org/viewtopic.php?id=310517
Offline
Post output of coredumpctl.
Write more about your system, hardware and drivers:
- Do you both use AMD CPU?
- Is installed microcode for CPU? What version?
- What AMD GPU do you have?
- Wayland or X server?
- What DE?
From part of logs I see that you both use amdgpu driver: MickeyRat had problems in game endless-sky and dingodoppelt in Wayland on KDE, but coredumpctl should shed more light on this.
@MickeyRat don't reinstall system yet, you have plenty of options: for example, if you have AMD CPU you can try downgrade ucode for it.
Last edited by xerxes_ (Yesterday 11:44:50)
Offline
Thanks for the reply. I have done a little more research. We are in agreement. At this point, I'm not certain a reinstall would fix it. The messages I posted are from endless sky but, I've gotten them from the brave browser as well; maybe other things. I had the problem for a while but, with the holidays, I didn't have time to look at it.
corredumpctl
TIME PID UID GID SIG COREFILE EXE SIZE
Fri 2025-11-28 21:53:25 EST 59804 1000 1000 SIGABRT present /usr/bin/endless-sky 251.4M
Fri 2025-11-28 22:31:05 EST 790 0 0 SIGABRT inaccessible /usr/lib/Xorg -
Fri 2025-11-28 23:30:53 EST 254124 1000 1000 SIGSEGV present /usr/bin/transgui 2.2M
Sat 2025-11-29 09:32:51 EST 787 0 0 SIGABRT inaccessible /usr/lib/Xorg -
Sat 2025-11-29 10:57:04 EST 3025 1000 1000 SIGSEGV present /usr/bin/transgui 1.9M
Sat 2025-11-29 14:40:16 EST 6871 1000 1000 SIGABRT present /usr/bin/endless-sky 249.1M
Sat 2025-11-29 18:44:58 EST 795 0 0 SIGABRT inaccessible /usr/lib/Xorg -
Sat 2025-11-29 23:42:17 EST 784 0 0 SIGABRT inaccessible /usr/lib/Xorg -
Sat 2025-11-29 23:42:20 EST 5988 1000 1000 SIGABRT present /usr/bin/endless-sky 207MCPU: AMD Ryzen 9 7900 12-Core
Microcode: amd-ucode 20251125-1
GPU: AMD/ATI Raphael - That's part of the processor. I don't have a graphics card.
X server
DE: Home built with openbox, tint2, some other things I'll list if necessary.
I'm beginning to think I should change the subject of this thread. It doesn't look like this is going to turn out to be a problem with the kernel.
Some cause happiness wherever they go; others whenever they go.
- Oscar Wilde
Offline
I'm pretty sure downgrading amd-ucode to 20251111 fixed it for me. I had no crashes since the downgrade.
My CPU is an AMD 7900X. I do have a dedicated graphics card (RX480) but it is not involved in this error.
KDE/wayland; X11 is not an option ![]()
coredumpctl didn't register, because there is no crash. I can still ssh into my machine and shut it down, though not completely.
Offline
@MickeyRat try downgrading linux-firmware-amdgpu to version 20251111-1. Not the meta linux-firmware package, but the specific linux-firmware-amdgpu package.
Offline
If the firmware downgrade doesn't fix it for you, try downgrading amd-ucode. That fixed it for me
Offline
Thank you both for replying. First, I tried downgrading amd-ucode. I still got core dumps but, it didn't blow me back to LightDM. Then I downgraded linux-firmware-amdgpu. That seemed to fix it but,I wanted to know whether both were necessary. So, I put amd-ucode back to 20251125-1 and so far it's working. I haven't gone this long without a lockup since this started happening.
I must confess I'm seeing another type of core dump in my journal but, that's related to bluetooth. I doesn't appear to be related to this issue. I want to make sure this is solved before tackling something else.
Last edited by MickeyRat (Yesterday 22:08:40)
Some cause happiness wherever they go; others whenever they go.
- Oscar Wilde
Offline
Same issue as https://bbs.archlinux.org/viewtopic.php?id=310497?
Offline
@loqs I wouldn't know for sure. A lot of that one sails over my head but, it sure could be. If I'm reading that thread right, it's been reported so it can be fixed.
Some cause happiness wherever they go; others whenever they go.
- Oscar Wilde
Offline
Does your system also have something similar to the following in the journal from when the issue occurs?
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: GCVM_L2_PROTECTION_FAULT_STATUS:0x00800932
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: Faulty UTCL2 client ID: CPF (0x4)
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: MORE_FAULTS: 0x0
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: WALKER_ERROR: 0x1
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: PERMISSION_FAULTS: 0x3
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: MAPPING_ERROR: 0x1
Nov 28 02:33:35 arch-framework-d kernel: amdgpu 0000:c3:00.0: amdgpu: RW: 0x0Offline
@loqs I can't find anything like that but, I'm not running a framework laptop either.
Some cause happiness wherever they go; others whenever they go.
- Oscar Wilde
Offline
It's probably this bug. It's been bisected and reverted upstream.
Offline
Please open an issue on https://gitlab.archlinux.org/archlinux/ … e/-/issues requesting the merge commit https://gitlab.com/kernel-firmware/linu … 9272dc054d or the three commits it contained https://gitlab.com/kernel-firmware/linu … c38f4cc947 https://gitlab.com/kernel-firmware/linu … de43ad632d https://gitlab.com/kernel-firmware/linu … b3adaa62dd be cherry-picked.
Offline
Pages: 1