Complete total freeze since 6.15 kernel

BigPandaPete · 2025-06-22 12:04:10

Since the kernel version 6.15 dropped I have a strange issue: system freezes off completely after 1-2 minutes since the beginning of the session.
I've tried clean reinstall with linux and linux-zen kernels but the issue remains. Going to linux-lts solves the issue.

A quick check of a log from the 'journalctl -b -1' also doesn't seem to help: I can't see any specific error from the kernel or anywhere else. It seems the system just hangs and becomes unresponsive without any error messages.
Is there maybe some known issues with the new version of the kernel when it comes to AMD cpu/gpu?

Just in case, my specs:

OS: Arch Linux x86_64
Host: MS-7E26 (1.0)
Kernel: Linux 6.12.34-1-lts (or 6.15.3.zen1-1)
Shell: bash 5.2.37
Display (LG Electronics 22"): 1920x1080 @ 60 Hz in 22" [External]
Display (M27Q): 2560x1440 @ 170 Hz in 32" [External] *
DE: GNOME 48.2
WM: Mutter (Wayland)
CPU: AMD Ryzen 7 9700X (16) @ 5.58 GHz
GPU 1: AMD Radeon RX 7900 XTX [Discrete]
GPU 2: AMD Radeon Graphics [Integrated]
Memory: 3.55 GiB / 60.40 GiB (6%)
Swap: 0 B / 4.00 GiB (0%)

Last edited by BigPandaPete (2025-06-22 18:33:24)

sveet_pickle · 2025-06-22 23:05:28

i have a similar problem thats resolved by switching to the lts kernel. I get boot loops on the login screen frequently as well on the latest that I assume are related. i see errors related to the gpu though when i run journalctl -b -1 -e

Shawesom · 2025-06-24 05:01:31

This has been happening to me since 6.15, LTS kernel is fine.
Specs: ASUS ROG Strix G513RM laptop
AMD ryzen 7 5900H cpu/apu
Nvidia 3060 maxq
Gnome 48.2
wayland (mutter)
32gb ram

anything you find or have learned since this post?

seth · 2025-06-24 06:53:04

There're numerous threads about various iterations of the 20250613 amdgpu firmware breaking things, but there's no indication for that in the journal provided by the OP - but it also ends w the initramfs (likely because of the freeze resulting in a hard reboot)
In addition there's https://bbs.archlinux.org/viewtopic.php?id=305632 and https://bbs.archlinux.org/viewtopic.php?id=305905 (mutter + xwayland specific freezes)

Avoid rebooting w/ the power button at all costs.
When the system "freezes" try to
1. switch the VT (ctrl+alt+f3)
2. ssh into it
3. frenetically press ctrl+alt+del to trigger a soft reboot
4. https://wiki.archlinux.org/title/Keyboa … el_(SysRq) (needs to be enabled *before* the freeze for alt+sysrq+reisub to work, sysrq is typically your print key)
This shall hopefully get us a journal hinting at what's going badly here.

grimler · 2025-06-25 15:36:39

I am experiencing a similar issue with AMD Radeon RX 9070 GPU, AMD Ryzen 9 9950X CPU and a MSI MEG x870e godlike motherboard.

It is very laggy, but not totally frozen. Pretty sure it started with linux-firmware update rather than kernel update for me, as @seth hinted at, If I connect a screen over display port it stops working in the middle of kernel logs. With hdmi it loses signal but comes back after a second or so.

I have removed quiet from cmdline and see that dmesg is filled with messages like:

[ 56.831192] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 57.145074] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 57.367166] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 57.685188] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 58.236962] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 58.457711] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 58.680017] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data
[ 58.905163] amdgpu 0000:03:00.0: [drm] *ERROR* dc_dmub_srv_log_diagnostic_data: DMCUB error - collecting diagnostic data

Does anyone have suggestions on what to run to perhaps get better logs? Will see if downgrading linux-firmware-amdgpu solves it..

Edit: seems my error is reported here: https://gitlab.freedesktop.org/drm/amd/-/issues/4238

Edit2: my error is fixed after upgrading linux-firmware to 20250613.12fe085f-9.

Last edited by grimler (2025-06-25 15:53:46)

seth · 2025-06-25 15:43:33

First -Syu, there have been confirmations of 20250613.12fe085f-9 fixing stuff™ (no guarantees, but worth a shot)

emfoli · 2025-06-26 16:19:14

I still didn't test the updated *-9 firmware packages but I'm also experiencing random freezes on a Zenbook S 14 after upgrading to kernel 6.15 series (although not that frequent -- twice after moving from 6.14 to 6.15). I did try the LTS kernel but -- for this specifc laptop -- the audio drivers only started working after linux 6.13 so I ended up downgrading back to Linux 6.14.9... I did run sevaral hardware tests after the freezes just to make sure Hardware was ok but nothing was caught. While running on series 6.14 kernels, I never experienced any issues.

emfol@zenbooks14-em
-------------------
OS: Arch Linux x86_64
Host: ASUS Zenbook S 14 UX5406SA_UX5406SA 1.0
Kernel: 6.14.9-arch1-1 (before, 6.15.3)
Uptime: 1 day, 23 hours, 36 mins
Packages: 1008 (pacman), 5 (flatpak)
Shell: bash 5.2.37
Resolution: 2560x1440
DE: GNOME 48.2
WM: Mutter
WM Theme: Adwaita
Theme: Adwaita [GTK2/3]
Icons: Adwaita [GTK2/3]
Terminal: tmux
CPU: Intel Ultra 7 258V (8) @ 4.800GHz
GPU: Intel Arc Graphics 130V / 140V]
Memory: 22579MiB / 31562MiB
Disk (/): 22G / 125G (19%)
Battery0: 83% [Charging]
Font: Adwaita Sans 11 [GTK2/3]

klutzer · 2025-06-26 20:07:49

@emfoli I'm having the very same issue, with the very same laptop. After upgrading, basically everyday the system freezes completely and if I use the LTS one, the sound doesn't work. Also, there are no hardware issues.

erico@erico-zenbook
--------------------------------------------------
OS: Manjaro 25.0.4 Zetar
Kernel: x86_64 Linux 6.12.34-1-MANJARO (before 6.15.3 as well)
Uptime: 18h 11m
Packages: 1272
Shell: zsh 5.9
Resolution: No X Server
DE: GNOME 48.2
WM: Mutter
WM Theme:
GTK Theme: adw-gtk3-dark [GTK2/3]
Icon Theme: Papirus-Maia
Font: Noto Sans 11
Disk: 145G / 2,7T (6%)
CPU: Intel Core Ultra 7 258V @ 8x 4.7GHz [38.0°C]
GPU: Mesa Intel(R) Graphics (LNL)
RAM: 17215MiB / 31567MiB

seth · 2025-06-26 20:23:15

Both of you have completely different HW than the OP, Shawesom and grimler
Actually, the common factor in this thread is gnome …

BigPandaPete · 2025-06-28 08:04:53

I've tested this with both GNOME and KDE: the issue remains in both cases. Concerning the avoidance of the hard reboot that @seth mentioned: unfortunately, power button is the only thing I can do - the system is completely unresponsive once it freezes, so no soft reboot nor switching VT (or anything else for that matter) is working. Most probably the issue is with how my amd gpu is handled by the new kernel - I'll try to change the card and test it in the near future.

seth · 2025-06-28 08:23:13

sysrq doesn't respond either?
If you keep "dmesg -w" running in a permanently visible terminal, can you catch some error messages before the freeze?

BigPandaPete · 2025-06-29 17:38:40

@seth I was going to check that but it seems the last update (linux-firmware 20250627-1 and linux-zen 6.15.4.zen2-1) fixed it for me. Spend couple of 1-2 hours long sessions now and the system worked fine. I'll keep checking for some time and report if the issue reappears.

seth · 2025-06-29 19:30:39

\o/
Please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.

Let's all hoe that AMD finally got their act together

emfoli · 2025-07-01 05:28:00

More than 4 days of uptime after downgrading to kernel 6.14.9 and no issues. During all these days, lots of coding, compiling, running containers, 4k video playback, screen share on Zoom, connecting and disconnecting to external monitors and a bit of gaming...
Since a new versions of the kernel and firmware are out, I'm gonna give them a try and share my experience. Wish me luck...

emfol@zenbooks14-em
-------------------
OS: Arch Linux x86_64
Host: ASUS Zenbook S 14 UX5406SA_UX5406SA 1.0
Kernel: 6.14.9-arch1-1
Uptime: 4 days, 5 hours, 56 mins
Packages: 1012 (pacman), 5 (flatpak)
Shell: bash 5.2.37
Resolution: 2880x1800
DE: GNOME 48.2
WM: Mutter
WM Theme: Adwaita
Theme: Adwaita [GTK2/3]
Icons: Adwaita [GTK2/3]
Terminal: tmux
CPU: Intel Ultra 7 258V (8) @ 4.800GHz
GPU: Intel Arc Graphics 130V / 140V]
Memory: 21709MiB / 31562MiB
Disk (/): 23G / 125G (20%)
Battery0: 63% [Discharging]
Font: Adwaita Sans 11 [GTK2/3]
Local IP: 192.168.68.67
Public IP: 179.199.108.15

klutzer · 2025-07-01 14:35:22

I'm testing with the last RC kernel (precisely 6.16.0-rc3-1) and it's going fine til now. No more freezes and no hardware issues as well (sound, BT, video, wireless, all good).

KenMacD · 2025-07-09 20:32:27

I was running in to similar on 6.15 kernel versions and found it was related to io_uring. Setting the sysctl `kernel.io_uring_disabled = 2` on boot fixed it for me. ymmv, but hope it helps.

loqs · 2025-07-09 20:36:48

KenMacD wrote:

Setting the sysctl `kernel.io_uring_disabled = 2` on boot fixed it for me

Do you still require that setting with 6.16-rc5 and if so is there an upstream bug report for the issue?

klutzer · 2025-07-10 14:31:34

The freeze just happened with me again on kernel 6.16. Trying the "kernel.io_uring_disabled = 2" config now.

klutzer · 2025-07-15 19:14:23

Didn't worked. Just got the freeze again with 6.16.0rc5-2

seth · 2025-07-15 20:31:52

The LTS kernel is still ok?
Does it also happen

OP wrote:

after 1-2 minutes since the beginning of the session

or more after a suspend?
x-ref, https://bbs.archlinux.org/viewtopic.php?id=306813

klutzer · 2025-07-15 20:41:04

@seth In my case it happens randomly after a few hours after turning on. I couldn't have found any pattern, and the "journalctl --boot=-1" doesn't help because it didn't point anything related there. Is it also important to say that the freeze is complete, I can't even Alt + PrtSc + REISUB on it.

seth · Yesterday 14:17:13

Sorry,

cat /mnt/var/lib/pacman/sync/core.db | curl -F 'file=@-' 0x0.st

klutzer · Yesterday 15:26:54

Sorry if it wasn't for me (it seems it is solved for @BigPandaPete, is that right?). But if was, here is the result: http://0x0.st/8kRq.bin

I'm also testing now with the updated Kernel 6.15 (x86_64 Linux 6.15.6-1-MANJARO) with power mode in "Performance" which I've seen it temporarily solves the freezing for some people.

seth · Yesterday 18:27:21

No, sorry - this was for an entirely different thread (I already wondered where I had posted that )

klutzer · Yesterday 18:45:04

No problem.

@BigPandaPete @emfoli have you guys solved the freezing issue?

Arch Linux

#1 2025-06-22 12:04:10

Complete total freeze since 6.15 kernel

#2 2025-06-22 23:05:28

Re: Complete total freeze since 6.15 kernel

#3 2025-06-24 05:01:31

Re: Complete total freeze since 6.15 kernel

#4 2025-06-24 06:53:04

Re: Complete total freeze since 6.15 kernel

#5 2025-06-25 15:36:39

Re: Complete total freeze since 6.15 kernel

#6 2025-06-25 15:43:33

Re: Complete total freeze since 6.15 kernel

#7 2025-06-26 16:19:14

Re: Complete total freeze since 6.15 kernel

#8 2025-06-26 20:07:49

Re: Complete total freeze since 6.15 kernel

#9 2025-06-26 20:23:15

Re: Complete total freeze since 6.15 kernel

#10 2025-06-28 08:04:53

Re: Complete total freeze since 6.15 kernel

#11 2025-06-28 08:23:13

Re: Complete total freeze since 6.15 kernel

#12 2025-06-29 17:38:40

Re: Complete total freeze since 6.15 kernel

#13 2025-06-29 19:30:39

Re: Complete total freeze since 6.15 kernel

#14 2025-07-01 05:28:00

Re: Complete total freeze since 6.15 kernel

#15 2025-07-01 14:35:22

Re: Complete total freeze since 6.15 kernel

#16 2025-07-09 20:32:27

Re: Complete total freeze since 6.15 kernel

#17 2025-07-09 20:36:48

Re: Complete total freeze since 6.15 kernel

#18 2025-07-10 14:31:34

Re: Complete total freeze since 6.15 kernel

#19 2025-07-15 19:14:23

Re: Complete total freeze since 6.15 kernel

#20 2025-07-15 20:31:52

Re: Complete total freeze since 6.15 kernel

#21 2025-07-15 20:41:04

Re: Complete total freeze since 6.15 kernel

#22 Yesterday 14:17:13

Re: Complete total freeze since 6.15 kernel

#23 Yesterday 15:26:54

Re: Complete total freeze since 6.15 kernel

#24 Yesterday 18:27:21

Re: Complete total freeze since 6.15 kernel

#25 Yesterday 18:45:04

Re: Complete total freeze since 6.15 kernel

Board footer