Got it. Okay, I’ll plan to run 6.6 for a couple of weeks to test even whether it was a driver regression as opposed to something else, and then if that works out okay I might try bisecting based on that and see what happens. My symptoms are a little different from yours (basically everything in the system is fine, just the internal display is hosed, and generally nothing in the log) but maybe there is some overlap in terms of the reason for the problem.
Yeah, but only the display is frozen, and I’d have to check if I could still connect a second monitor via DP the next time it happens. I also can’t tell if maybe just the output is frozen and I could technically switch the tty and enter stuff there blindly. ssh still works, so I can use the sysreq key sequence, pass it as a little bash snippet to the system and sometimes I even can get back to a working system by unloading and reloading the amdgpu driver module, though not reliably. Also, if an audio source was outputting at the time, audio still works and I can still use the play/pause button on F5 to control it. So just the GPU driver itself seems to have an issue.
Of course a cool workaround would be to have a script being triggered by that log line that will unload the driver module and somehow just transition over to software rendering - llvmpipe is quite capable after all, but no idea if such transition is even possible, especially without my Gnome session being closed like it is when reloading the driver module.
Oh. Yeah, that sounds very similar to my symptoms. I also did get messages in the log sometimes, it was just rare (usually silent). So maybe it is the same.
All the attempts at various workaround kernel parameters have now failed again, so now I’m onto mucking with kernel versions, let’s see what happens.
One thing I’ve noticed that may be a thing or just coincidence is that since 6.16 - and I’d say especially 6.16.3 I had this quite a few times happen to me, but it hasn’t happened yet again since 6.16.5. If you can reproduce this, you might not have to dig that deep.
Oh, interesting. Sure, I’ll try it. I’m updating to 6.16.5 and let’s see. It only really happens about once a week for me, so it’s pretty tough to test whether or not something has “solved” it, but if 6.16.5 resolves things then I’d be happy with that.
So I’ve got a bit of news. It just happened again on 6.16.6. So it’s not fully fixed. I couldn’t connect an external monitor via DP, but that was quite obvious. But also thermals don’t really look like they could cause this:
cros_ec-isa-000c
Adapter: ISA adapter
fan1: 0 RPM
fan2: 0 RPM
ambient_f75303@4d: +37.9°C
charger_f75303@4d: +35.9°C
apu_f75303@4d: +36.9°C
cpu@4c: +39.9°C
gpu_amb_f75303@4d: -0.1°C
gpu_vr_f75303@4d: -0.1°C
gpu_vram_f75303@4d: -0.1°C
gpu_amdr23m@40: -0.1°C
ucsi_source_psy_USBC000:004-isa-0000
Adapter: ISA adapter
in0: 0.00 V (min = +0.00 V, max = +0.00 V)
curr1: 0.00 A (max = +0.10 A)
spd5118-i2c-2-50
Adapter: SMBus PIIX4 adapter port 0 at 0b00
temp1: +41.5°C (low = +0.0°C, high = +55.0°C)
(crit low = +0.0°C, crit = +85.0°C)
ucsi_source_psy_USBC000:002-isa-0000
Adapter: ISA adapter
in0: 5.00 V (min = +5.00 V, max = +5.00 V)
curr1: 0.00 A (max = +1.50 A)
nvme-pci-0200
Adapter: PCI adapter
Composite: +35.9°C (low = -5.2°C, high = +89.8°C)
(crit = +93.8°C)
acpitz-acpi-0
Adapter: ACPI interface
temp1: +37.8°C
temp2: +35.8°C
temp3: +36.8°C
temp4: +39.8°C
k10temp-pci-00c3
Adapter: PCI adapter
Tctl: +40.9°C
iwlwifi_1-virtual-0
Adapter: Virtual device
temp1: +40.0°C
amdgpu-pci-c100
Adapter: PCI adapter
vddgfx: 810.00 mV
vddnb: 761.00 mV
edge: +38.0°C
PPT: 7.02 W (avg = 6.17 W)
sclk: 800 MHz
spd5118-i2c-2-51
Adapter: SMBus PIIX4 adapter port 0 at 0b00
temp1: +39.8°C (low = +0.0°C, high = +55.0°C)
(crit low = +0.0°C, crit = +85.0°C)
ucsi_source_psy_USBC000:003-isa-0000
Adapter: ISA adapter
in0: 0.00 V (min = +0.00 V, max = +0.00 V)
curr1: 410.00 mA (max = +0.00 A)
ucsi_source_psy_USBC000:001-isa-0000
Adapter: ISA adapter
in0: 0.00 V (min = +0.00 V, max = +0.00 V)
curr1: 0.00 A (max = +0.10 A)
BAT1-acpi-0
Adapter: ACPI interface
in0: 17.22 V
curr1: 710.00 mA
So my only theory why this can happen and can happen multiple times in a row unless you “start up your device slowly” (i.e. wait before login, before opening any program etc a minute or two) is out the window.
Yeah, I’ve done that multiple times. It’s just such an intermittent problem that I’ll think it is fixed because of some change, only for a week and a half to go by and then for it to crop up again.
Thanks for the update. I first ran NixOS 202411 on this machine for some decent length of time (at least subjectively speaking) and didn’t run into the issue, so I may go back to the plan of going back to kernel 6.6 which I think was what was on it at that time, and see if I can go for a couple of weeks with no issue.
I can confirm this issue on Bazzite (Gnome, Wayland, VRR enabled) on Kernel versions 6.15 and 6.16. Happens constantly.
This issue makes the 16 completely unusable for me. Freezes don’t go away with any kernel parameters and don’t go way on their own either, plus they start right after logon. On Fedora 42 with the latest kernel + GNOME.
Yeah, sounds about right. I have rolled back to kernel 6.6.101, and it went a few weeks without me seeing the issue at all. I tried 6.16.5 for testing today and it froze instantly. Some configs make it much less frequent, but some configs make the laptop pretty much useless.
I think it might be worth trying out an older kernel and seeing if the issue goes away for you as well. It’s so intermittent under some configs that it’s hard to say when it is or isn’t completely resolved, but I’ve had good experience with 6.6 and I’ve heard someone else say that 6.8 also resolved it for them.
Thanks for the suggestion, I’ll try that!
Edit: Switching to kernel 6.8 resolved my issue. For anyone having trouble with this in the future I recommend trying that, at least until this bug is resolved.
Edit 2: The bug still occurs, but the screen can be unfrozen by suspending. Still an improvement.
BIOS update 3.07, released yesterday, appears to fix this, even while running the newer kernel.
Oh, interesting, thank you for the tip. I will say that a lot of things make the problem less frequent without fixing it completely; I started setting 2 weeks as the minimum time of not seeing the issue before I started saying tentatively that it actually was resolved.
On my side, one of the AMD developers gave me a patch to try, Framework support offered a new mainboard RMA, I decided to order more memory (I was using a single 8GB stick for whatever weird reason, which if it was related to GPU memory allocation might potentially have been related I guess), and also now I have a new BIOS version to try. I might honestly just try all four things, and if it’s fixed, I don’t really have to care about which one of them it was that actually was responsible.
(I am a little bit tempted to try the kernel patch in isolation first, at least for a few days, just to provide useful feedback. Even if the feedback includes caveats about a BIOS update also or something.)
On my side, it was pretty much the opposite. Save for rolling back the kernel, a lot of the proposed fixes (e.g. setting debug mask) didn’t make it less frequent at all. Then one BIOS update stops it completely 
On the other hand if I was running something like arch I don’t know when I would have known there was an update available. It was only because Fedora automatically fetched it that I knew. I hope others can discover the patch reasonably quickly, because I suspect any regular end-user would have already switched to windows in this time frame.
Save for rolling back the kernel, a lot of the proposed fixes (e.g. setting debug mask) didn’t make it less frequent at all. Then one BIOS update stops it completely
Ha. Yeah, who knows, I can only report what happened for me.
I hope others can discover the patch reasonably quickly, because I suspect any regular end-user would have already switched to windows in this time frame.
Yeah pretty much. I do love the laptop overall, I am not trying to sound like I am super-aggrieved about it since it works most of the time, but certainly the little malfunctions that seem to come with it and have had no resolution so far make it a less positive experience.
Also imagine that instead of fighting with it for a week or so and then getting resolution (if that is what happened, I am just guessing from your user not having posted about it before)… it had been months and months of this, with two separate problems where sometimes a partial resolution to one, makes the other one more likely to happen. That has been my experience so far. 
I will add - the BIOS update didn’t completely fix it. Actually, the bug still occurs if I suspend the laptop, but at least it isn’t unusable anymore. I don’t really use suspend anyway, though it is a bit annoying.
EDIT: Well, scratch that, it still breaks occasionally. Honestly, for this price with a bug that breaks both suspend and will freeze the laptop after like an average of 2 hrs use, it’s really not worth it to buy the 16.
Have you created a support ticket?
This sounds more like it could be faulty hardware than anything else. But it seems you only have this issue on Fedora, right?
I am on Fedora 43 and have none of these issues on Kernel: Linux 6.17.6-300.fc43.x86_64
Have you tried booting into Windows and just testing if its happening there? That would help verify if its hardware or software related.
Maybe also boot a live usb of Fedora and see if it reproduces there.
I get that this is frustrating, but I dont think you are hitting the bug, and if you are, its not really on framework’s end. Its on the developers for the distro you are running. Have you tried to create a bugzilla for Fedora? Fedora Bugzilla link
Report in there the issue, the versions you are running, and collect the journalctl logs that it requests when you file.
Reconfirming the bug on Bazzite 43 (GNOME 49), kernel version 6.17. BIOS version 3.07.
Pressing the power button to suspend the system and then pressing the power button to turn it on again fixes the issue.
It might actually be a hardware problem (?) I don’t just experience it on Fedora, or I would just switch distros. The bug I’m having is a bit different from what other people have reported. I’ll look into it and try it on windows.