There was an update (akmod-nvidia-470xx-470.141.03-3) this morning in the rpmfusion non-free repo that fixed this for me.
I can now boot into kernel 6.0.5 with the nvidia driver. $ sudo dnf update akmod-nvidia-470xx --enablerepo=rpmfusion-nonfree-updates-testing
$ rpm -qi akmod-nvidia-470xx
Name : akmod-nvidia-470xx
Epoch : 3
Version : 470.141.03
Release : 3.fc37
Architecture: x86_64
Install Date: Mon 31 Oct 2022 01:09:10 PM EDT
Group : Unspecified
Size : 95109
License : Redistributable, no modification permitted
Signature : RSA/SHA256, Sun 30 Oct 2022 06:58:03 PM EDT, Key ID 6a2af96194843c65
Source RPM : nvidia-470xx-kmod-470.141.03-3.fc37.src.rpm
Build Date : Sun 30 Oct 2022 06:13:05 AM EDT
Build Host : buildvm-07.virt.rpmfusion.net
Packager : RPM Fusion
Vendor : RPM Fusion
URL : https://www.nvidia.com/
Summary : Akmod package for nvidia-470xx kernel module(s)
Description :
This package provides the akmod package for the nvidia-470xx kernel modules.
What rpmfusion repo are you using. dnf repolist will tell us that.
Also, what akmod-nvidia package is already installed? dnf list installed akmod-nvidia and post the entire command and result.
There are 2 problems there if your system requires the 470xx drivers.
You have the rpmfusion-nonfree-nvidia-driver repo enabled which in the past did not contain the 470 drivers and which may conflict with the rpmfusion-nonfree repo when trying to install the proper driver.
As noted, you seem to not have the nvidia packages installed. We can verify which nvidia packages are actually installed with dnf list installed \*nvidia\*
yes, assuming the repo that holds the new rpm is enabled (i.e. if its in the testing repo then updates-testing is enabled). The ‘dnf clean expire-cache’ will ensure that the list of packages is using to the latest list.
@vk2bea
you are giving instructions for F37 while the OP is on F36. Similar but not exactly the same. Updates-testing is for F37 and is the fedora repo not the rpmfusion repo, which would be rpmfusion-nonfree-updates-testing.
@heikonion
It would appear you have the correct version of the rpmfusion drivers installed but it failed to properly build the modules.
I suggest you
Reboot to the 5.19.16 kernel
remove the 6.0.5 kernel sudo dnf remove kernel*6.0.5*
reboot
reinstall the latest kernel sudo dnf upgrade --refresh
Wait at least 5 minutes then check the modules to see if they were built. dnf list installed kmod-nvidia* which now should show an entry for the 6.0.5 kernel.
if the kmod-nvidia-470xx-6.0.5* package now shows up then reboot to the new kernel and the drivers should properly be loaded.
hold down ‘shift’ or ‘esc’ while booting from the bios screen and it should bring up the grub menu. Otherwise we might need to use efibootmgr to select that.
We have to force the system to build the nvidia driver modules. The removal, and install of the kernel seems to me the simplest way to ensure the modules are built properly.