AMD Ryzen 7040 Series BIOS 3.03 and Driver Bundle Release

I was experiencing a similar issue on Fedora 39 KDE as well. Looks like we’re not alone - there is an as yet unresolved issue in the kernel amdgpu driver affecting AMD APUs.

Since adding kernel parameter amdgpu.sg_display=0 in /etc/default/grub and regenerating grub config ( sudo grub2-mkconfig -o "$(readlink -e /etc/grub2.conf)" ), I have not yet experienced further white flickering.

Hopefully there will be a permanent fix soon.

I just received a Batch 2 laptop today (20Oct23). The BIOS is still 3.02 and it suffers from the reported problem that, after boot, it launches with X11 but logging out and back in switches to Wayland. (Fedora 39 beta, fully updated.)


I renamed the title “AMD Ryzen 7040 Series BIOS 3.03 BETA release” to “AMD Ryzen 7040 Series BIOS 3.03 Beta release”. Because the “beta” is not the abbreviation. It also aligns with the other past BIOS threads that Kieran created such as the following threads. I hope you still like it.

1 Like

Forgive my naïveté, are these BIOS issues effecting the Microsoft people as well? I see a lot of Linux OS mentions but haven’t noticed any Microsoft OS mentions.

One user had several BSOD even after multiple fresh Windows installs but it’s unknown if that’s due to BIOS 3.02 or a defective board.

This is a single data point so take it with a grain of salt, but I’ve had no issues with Windows 11 so far on a clean install even with a technically-unsupported AX210 NIC.

1 Like

I don’t have a lot of recent experience with PCs as I’ve spent the last decade on Apple devices. I’m sincerely hoping I don’t have a lot of issues that will require technical expertise at a level which I am no longer proficient

You don’t have to do it alone! You’ll be troubleshooting together with Framework’s highly knowledgeable support team

That’s good to know, thank you!

This is an amdgpu linux kernel driver issue that can be sort of mitigated by a simple extra kernel option for the time being.

Look here: this thread

Is there a chance we might be able to get a (tentative) timeline for a release of the 3.03 update? As in, sometime this month still, mid-November, Q1 2024 etc?

I understand this doesn’t affect Windows users much, just Linux users on newer kernels. I also realise there’s a workaround by running an LTS-derived kernel on some distributions, but that’s not ideal for many folks. Even though security updates are normally backported to LTS kernels, many driver improvements and additional hardware support are typically not, not even by vendors, unless it specifically affected a customer with a support contract (which wouldn’t happen on Fedora for example).

The answer is …

But for sure we seem to not have the same understanding of “soon” ahahah.
Expecially when we’re expecting something. Let’s wait a little more. The bios had to be released at the same time than batch 2 but did not make it. It makes me think that they’ve catched some last minute bugs just before the expected release.

Very possible, and I’m entirely fine with it. If it takes another month, it takes another month. That’s not the issue. But it would be nice to get some clarity on that.

1 Like

Thanks, it’s been a couple of days and this seems to have fixed the issue for me!


Sounds promising especially after the long testing period by framework. Thank you all for your work!

1 Like

I believe it’s actually older kernels that are affected. Notably latest stable (6.5.8 as of writing) seems unaffected according to what people have claimed on the forum.

Does that improve performance, or temperature? (i.e. What’s the rational / driver for the PL table modification?)

Maybe reviewers can do some post-BIOS-update performance benchmarking again?

(Remotely related: Asus Ally went through a rollercoaster period with BIOS updates impacting performance)

This is to improve performance if the mainboard is operating without a battery. It should have little/no impact when the system is operating with a battery in a normal laptop configuration.


Thank you for the prompt response. So this is related to the standalone / in-a-case use case. Neat that it has those users covered.

Regarding Kernel 6.5, is there a minimum patch version dependency? (e.g. 6.5.7-300 from the Fedora side of things?)