Framework 13 Pro Fedora 44 Frequent System Freezes

Framework Laptop 13 Linux
1

Which Linux distro are you using?

Fedora 44 Workstation Gnome 50 updated yesterday.

Which kernel are you using?

Linux fedora 7.2.5-200.fc44.x86_64 #1 SMP PREEMPT_DYNAMIC Fri Sep 11 15:11:05 UTC 2026 x86_64 GNU/Linux

Which BIOS version are you using?

dmidecode 3.7

Getting SMBIOS data from sysfs.
SMBIOS 3.3.0 present.

Which Framework Laptop 13 model are you using? (AMD Ryzen™ AI 300 Series, AMD Ryzen™ 7040 Series, Intel® Core™ Ultra Series 1, 13th Gen Intel® Core™ , 12th Gen Intel® Core™, 11th Gen Intel® Core™)

Framework Laptop 13 Pro (Batch 7)

Version: AMD Ryzen AI 7 350 w/ Radeon 860M

The problem I’m having is that ever so often the system just freezes. There does not appear to be any uniform cause except I just clicked on something like a link in firefox, or launched a video, or whatever. This happens at least once a day, but no more than a few times a day. The only way out is to hold down the power button for (approx) 10 seconds, and the power on again.

How do I track the problem if the computer freezes? Is there anything I can do while it is frozen? Is there anything I can look at after rebooting?

2

Do you see anything in your historical logs about page flip timeouts by chance?

If so; 7.3-rc4 has the most probable fix.

3

I have no idea how to find the historical logs of which you write.

I tried updating fedora so now I have:

Linux fedora 7.2.7-200.fc44.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Sep 21 18:25:57 UTC 2026 x86_64 GNU/Linux

Is this not the OS that Framework supports? So maybe fedora is not the OS I should be using, and I need to go back to Debian.

4

It’s got nothing to do with fedora vs Debian. There is a regression in 7.2 series that I suspect you’re hitting.

Please share journalctl -k -b-1 output after you hit the issue. This will get logs from previous boot.

5

It just froze again, so I have the output frm journalctl -k -b-1. It’s over 2400 lines. How do I post it?

An error occurred: Body is limited to 32000 characters; you entered 261710.

6

Just split your text into two pieces, and then paste each one separately, in this thread. For each one, select all, then hit the </> icon, so they are readably formatted.

7

Just upload it to a bug in a bug tracker and link here. Kernel bugzilla, fedora bugzilla, drm/AMD gitlab. I don’t care where.

1 Like
8

You mean 9 pieces?

9

I don’t use bug trackers. How about: Sep 23 18:19:06 fedora kernel: Linux version 7.2.7-200.fc44.x86_64 (mockbuild@4b - Pastebin.com

10

What is CHTTPClientThre?

This seems to be software causing a problem.

11

Here?

I had steam client running earlier, but steam wasn’t running when it froze.

12

So, 64 bit atomic across 2 cache lines in a 32 bit application.

I feel that this is something a compiler should be able to catch.

13

The split lock is apparently a red herring.

The split-lock logged by the Steam Client (CHTTPClientThread) are most likely not causing the freeze.

Those log messages stopped when I exited the steam client, and the system froze about a half hour later.

Is this final message in the log relevant?

Sep 24 16:39:59 fedora kernel: kernel BUG at lib/list_debug.c:32!

14

It’s hard to know for sure if that’s the cause of the freeze because you certainly have kernel memory corruption right after it.

Sep 24 16:39:59 fedora kernel: list_add corruption. prev->next should be next (ffff8b8e5e58ee30), but was 0000000000000100. (prev=ffff8b8e416d6a00).Sep 24 16:39:59 fedora kernel: ------------[ cut here ]------------Sep 24 16:39:59 fedora kernel: kernel BUG at lib/list_debug.c:32!

Fedora has a system to catch this kind of stuff (kdump/kcrash). I have no idea if it’s enabled by default, but my understanding is it lets you generate bug reports from these kinds of crashes and will capture what’s needed to analyze them (the vmcore).

15

The thing about list corruption is you generally don’t know when it started unless you tear apart the vmcore with a debugger. It could have been way earlier in the boot.

16

Ah! I misread the scale of the pieces. OK, instead a GitHub Gist or a Pastebin should do it.

17

So is there smething I can actually do to find out what’s causing this? Or should I just save often and hope for the best?

18

Correction:

Vendor: INSYDE Corp.
Version: 04.02
Release Date: 06/09/2026
ROM Size: 32 MiB

Update:

Linux fedora 7.2.7-200.fc44.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Sep 21 18:25:57 UTC 2026 x86_64 GNU/Linux

19

You can send in a bug report with the vmcore if you have one. If you don’t have one, try to turn the service on and send one up.

If you don’t want to send a bug report (you said previously you didn’t want to) another option is you can also do a kernel checkout and use an LLM to analyze your vmcore locally to see if it can figure out the problem (and potentially generate a patch).

20

Sep 24 16:39:59 fedora kernel: list_add corruption. prev->next should be next (ffff8b8e5e58ee30), but was 0000000000000100. (prev=ffff8b8e416d6a00).

Note the 0x100.
Why would the value be 0x100.
Turns out 0x100 is defined as LIST_POISON1.
Set to that when entries are deleted.
list_add() is not thread safe, so my guess is whatever is calling it, might be forgetting to lock it first.