unfixable, by the way. but you may try X11
this post was submitted on 29 Oct 2025
18 points (95.0% liked)
Linux
10210 readers
488 users here now
A community for everything relating to the GNU/Linux operating system (except the memes!)
Also, check out:
Original icon base courtesy of lewing@isc.tamu.edu and The GIMP
founded 2 years ago
MODERATORS
Ah shit. This is it.
Just my luck that GNOME 49 just dropped support for X11. I guess if it's unbearable I can downgrade.
At least now I know what to watch. Thank you!
Whoa, looks like your home directory or user got orphaned somehow during the upgrade.
- Open a console terminal (ALT+CTRL+F3) and try to login there (it may throw errors)
- Run
whoand see what the output is - Run
sudo su -u, enter your password, thenls -lh /home
What's the output?
who:
user seat0 2025-10-29 16:47
user tty3 2025-10-29 16:47
user seat0 2025-10-29 16:34
user tty2 2025-10-29 16:34
ls - lh /home:
totalt 0
drwx-------. 1 user user 794 okt. 29 14:55 user
This was done before the computer had frozen, I'm not sure if that spoils it.
Check if systemd has any clues:
journalctl -b > systemd-logs.txtgrep -C3 -i "error\|fail|warn\|fatal\|can't\|cannot"systemd-logs.txt
I will run this next time it crashes (if it happens again) - after freezing several times in a row earlier it has been steady the last couple of hours, so the last crash is outside of the period of the log that I see now.
Thanks!
I'm having the same issue. I've seen it might not just be an AMD issue, it could be a GNOME issue, and it might be a wayland issue. Its driving me crazy. Some users said they were able to fix it by setting the kernel parameter amdgpu.dcdebugmask=0x10 which disables the Panel Self Refresh which is supposed to help with battery life. Someone said amdgpu.runpm=0 setting was also required (something to do with suspending the gpu). These didnt work for me and neither did switching desktop environments.
Heres a link to the kernel parameter discussion, and a guide I used.
The problem I've had with trying to debug this is that the journal logs are missing the last 5 minutes or so before the crash. I tried writing a script to log every half second, but had the same problem with missing logs. Could this indicate something? ctrl alt f3 doesn't work either when it freezes. Ran hardware tests and those all passed, BIOS and firmware are all up to date..
Considering some distro hopping.
It seems the running application remains responsive, but not anything else.
I was running RStudio on an external monitor once when it froze, and I could keep using the window. Even when I used touch gestures to "zoom" the window out, I remained in control over it and could execute code in the smaller window. But I could not interact with GNOME at all, including changing to another program on the same screen.
Another time it froze as I wrote a PieFed comment, and I pressed tab and enter a few times. The comment was published so I could continue where I left off.
First time in almost 20 years of Linux usage I'm encountering real problems, and I'm much too primitive to do anything clever about it.
Does it seem to be a Fedora problem?
I'm not sure if this is a Fedora problem, but this is a problem on a fresh install for me, so before I get too set up, I figured I can try something else. Your experience sounds a little different than mine though. I imagine an update will come around to fix this eventually, but maybe those kernel parameters will work for you?
I'll give it a shot, thanks!
I would try first to boot with the last kernel you were using before the upgrade.
Thanks! I didn't even realize this was an option - very useful to know! Will check it out if the crashes keep up. :)
@cabbage Might not be related, but I was triggered by your "flip_done timed out"... See:
https://community.frame.work/t/responded-linux-recent-update-is-causing-some-sort-of-stalling-freezing-in-pop-os/48537/18?u=florisnielssen
- Which window manager are you using?
- What typically 'fixed' it for me was switching to terminal ([CTRL]+[ALT]+[F3]) and afterwards switching back to the GUI again ([CTRL]+[ALT]+[F2]). It took a minute, but allowed me to resume my work.
I had issues on Pop_OS in combination with Wayland.
That sounds familiar - I will look into this if the crashes return (it has been stable for a couple of hours now). Going wild with screen brightness and night light does not seem to trigger anything.
I'm on Wayland and pretty vanilla Fedora with GNOME.
@cabbage For me, it also depended if I was working with a 2nd screen attached or only my laptop.
With only the laptop it happened more often.
The 2nd screen never froze, but sometimes the laptop screen would with the 2nd screen still working fine.
I had problems while an external monitor was connected on Fedora 42, but as I don't use it often I figured I could live with it. Now it seemed to have gotten worse. Once the second monitor was working but GNOME was acting weird. So yes, that absolutely does seem related.