If it helps, the laptop has two GPUs, one of them is an AMD GPU and the other one is this Nvidia 5070 Ti.
it could be a driver bug.
You could try to remove the driver and reinstall again, otherwise maybe go back to the previous version.
pls run
lspci | grep -i -E 'vga' | cut -b1-7 | xargs -i lspci -vnnks {} | grep -v "<access denied>"
Thanks,
This is the output of lspci | grep -i -E 'vga' | cut -b1-7 | xargs -i lspci -vnnks {} | grep -v "<access denied>"
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GB205M [GeForce RTX 5070 Ti Mobile] [10de:2f18] (rev a1) (prog-if 00 [VGA controller])
Subsystem: AIstone Global Limited Device [1d05:6054]
Physical Slot: 0
Flags: bus master, fast devsel, latency 0, IRQ 102, IOMMU group 16
Memory at f8000000 (32-bit, non-prefetchable) [size=64M]
Memory at f800000000 (64-bit, prefetchable) [size=16G]
Memory at fc00000000 (64-bit, prefetchable) [size=32M]
I/O ports at f000 [size=128]
Expansion ROM at fc000000 [disabled] [size=512K]
Kernel driver in use: nvidia
Kernel modules: nouveau, nvidia_drm, nvidia
07:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Granite Ridge [Radeon Graphics] [1002:13c0] (rev d4) (prog-if 00 [VGA controller])
Subsystem: AIstone Global Limited Device [1d05:5004]
Flags: bus master, fast devsel, latency 0, IRQ 53, IOMMU group 22
Memory at fc10000000 (64-bit, prefetchable) [size=256M]
Memory at fc20000000 (64-bit, prefetchable) [size=2M]
I/O ports at d000 [size=256]
Memory at fc500000 (32-bit, non-prefetchable) [size=512K]
Kernel driver in use: amdgpu
Kernel modules: amdgpu
I have removed and reinstalled it many times without any luck I’m afraid.
I need to find a way to go back to the previous version.
I removed everything Nvidia related using:
sudo dnf remove \*nvidia\*
And now I’m installing it again as a last test using
sudo dnf install nvidia-open
EDIT: no difference at all.
don’t remove ’ nvidia-gpu-firmware.noarch’ it’s needed by the nouveau driver, that is your current failback
it looks like the repo rpmfusion-nonfree-nvidia-driver was also updated.
So you have to download the rpm files from koji.rpmfusion.org
don’t forget to recreate the file in /etc/rpm/
1 Like
Thanks, installed the nvidia-gpu-firmware.noarch back again.
I’ll try to download the rpm files and get it sorted as this is my work laptop 
Thanks, I’ll try this again.
Nope, no luck at all. Going to install the previous version.
ok this looks like a driver bug,
don’t download the kmod-nvidia rpm you don’t it, only the akmod-nvidia rpm
1 Like
Thanks for you help mate.
Reinstalled 570 and everything is back to normal, phew!
Thank you ever so much. I need to blacklist it somehow in the repo files so it doesn’t get installed until it’s fixed.
you can add a dnf versionlock or you can add an exclude with
dnf config-manager setopt exclude=
man dnf-versionlock
man dnf-config-manager
1 Like
Thanks Mark,
Locked akmod-nvidia for the time being.
You’re a big life saver.
Hi all, today i tried to update my Silverblue, but i’ll get this warning about bad signature:
~$ rpm-ostree upgrade
2 metadata, 0 content objects fetched; 788 B transferred in 0 seconds; 0 Bytes content written
Checking out tree 02b4d18... done
Enabled rpm-md repositories: fedora-cisco-openh264 updates fedora rpmfusion-free-updates rpmfusion-free rpmfusion-nonfree-updates rpmfusion-nonfree copr:copr.fedorainfracloud.org:phracek:PyCharm google-chrome rpmfusion-nonfree-nvidia-driver rpmfusion-nonfree-steam updates-archive
Importing rpm-md... done
rpm-md repo 'fedora-cisco-openh264' (cached); generated: 2024-08-21T16:04:02Z solvables: 3
rpm-md repo 'updates' (cached); generated: 2025-06-09T02:30:52Z solvables: 16452
rpm-md repo 'fedora' (cached); generated: 2025-04-09T11:06:59Z solvables: 76879
rpm-md repo 'rpmfusion-free-updates' (cached); generated: 2025-06-05T23:35:46Z solvables: 64
rpm-md repo 'rpmfusion-free' (cached); generated: 2025-04-12T09:12:27Z solvables: 358
rpm-md repo 'rpmfusion-nonfree-updates' (cached); generated: 2025-06-05T23:50:08Z solvables: 28
rpm-md repo 'rpmfusion-nonfree' (cached); generated: 2025-04-12T09:32:45Z solvables: 218
rpm-md repo 'copr:copr.fedorainfracloud.org:phracek:PyCharm' (cached); generated: 2025-05-20T06:03:29Z solvables: 5
rpm-md repo 'google-chrome' (cached); generated: 2025-06-09T10:40:07Z solvables: 4
rpm-md repo 'rpmfusion-nonfree-nvidia-driver' (cached); generated: 2025-06-06T06:08:02Z solvables: 17
rpm-md repo 'rpmfusion-nonfree-steam' (cached); generated: 2025-05-23T11:17:51Z solvables: 1
rpm-md repo 'updates-archive' (cached); generated: 2025-06-09T03:01:32Z solvables: 19245
Resolving dependencies... done
Will download: 1 package (133,5 MB)
Downloading from 'rpmfusion-nonfree-updates'... done
Importing packages... done
error: importing RPMs: Paket xorg-x11-drv-nvidia-cuda-libs-3:575.57.08-1.fc42.x86_64 kann nicht überprüft werden und repo rpmfusion-nonfree-updates ist GPG aktiviert: /var/cache/rpm-ostree/repomd/rpmfusion-nonfree-updates-42-x86_64/packages/xorg-x11-drv-nvidia-cuda-libs-575.57.08-1.fc42.x86_64.rpm could not be verified.
/var/cache/rpm-ostree/repomd/rpmfusion-nonfree-updates-42-x86_64/packages/xorg-x11-drv-nvidia-cuda-libs-575.57.08-1.fc42.x86_64.rpm: PRÜFSUMME: SIGNATURE: NICHT OK
it looks like the cuda-libs package has a bad signature.
glad, the stable 570 driver works.
btw. found this, it’s exactly the same error, confirmed by NVIDIA
This doesn’t exactly look good for me either with a RTX2060 in Fedora 41 :
Jun 9 18:46:22 fed-obs kernel: NVRM: loading NVIDIA UNIX Open Kernel Module for x86_64 575.57.08 Release Build (dvs-builder@U22-I3-H04-01-5) Sat May 24 07:03:13 UTC 2025
Jun 9 18:46:22 fed-obs kernel: NVRM: s_executeBooterUcode_TU102: Booter failed with non-zero error code: 0x2
Jun 9 18:46:22 fed-obs kernel: NVRM: s_executeBooterUcode_TU102: Booter failed with non-zero error code: 0x2
Jun 9 18:46:22 fed-obs kernel: NVRM: kgspExecuteBooterLoad_TU102: failed to execute Booter Load: 0xffff
Jun 9 18:46:22 fed-obs kernel: NVRM: kgspExecuteBooterLoad_TU102: failed to execute Booter Load: 0xffff
Jun 9 18:46:22 fed-obs kernel: NVRM: kgspBootstrap_TU102: failed to execute Booter Load (ucode for initial boot): 0xffff
Jun 9 18:46:22 fed-obs kernel: NVRM: kgspBootstrap_TU102: failed to execute Booter Load (ucode for initial boot): 0xffff
Jun 9 18:46:22 fed-obs kernel: NVRM: _kgspBootGspRm: unexpected WPR2 already up, cannot proceed with booting GSP
Jun 9 18:46:22 fed-obs kernel: NVRM: _kgspBootGspRm: unexpected WPR2 already up, cannot proceed with booting GSP
Jun 9 18:46:22 fed-obs kernel: NVRM: _kgspBootGspRm: (the GPU is likely in a bad state and may need to be reset)
Jun 9 18:46:22 fed-obs kernel: NVRM: _kgspBootGspRm: (the GPU is likely in a bad state and may need to be reset)
Jun 9 18:46:22 fed-obs kernel: NVRM: RmInitAdapter: Cannot initialize GSP firmware RM
Jun 9 18:46:22 fed-obs kernel: NVRM: RmInitAdapter: Cannot initialize GSP firmware RM
Jun 9 18:46:22 fed-obs kernel: NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x62:0x40:1941)
Jun 9 18:46:22 fed-obs kernel: NVRM: GPU 0000:01:00.0: RmInitAdapter failed! (0x62:0x40:1941)
Jun 9 18:46:22 fed-obs kernel: NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
Jun 9 18:46:22 fed-obs kernel: NVRM: GPU 0000:01:00.0: rm_init_adapter failed, device minor number 0
What’s worse is that it seems I can’t revert back to akmod-nvidia 570 : trying to downgrade will go to akmod-nvidia 560 (I could still try that)
No issues here with my RTX 2060 on F42.
“Have you tried turning it off and on again?” 
Otherwise you could try the closed source driver instead
sudo sh -c 'echo "%_without_kmod_nvidia_detect 1" > /etc/rpm/macros.nvidia-kmod'
sudo akmods --kernels $(uname -r) --rebuild
modinfo -l nvidia should report ‘NVIDIA’ instead of ‘Dual MIT/GPL’
Reboot.
Seriously? It is described in this very thread where the 570 drivers can be downloaded…
Hi Mark and thanks for you reply. You’re using akmod-nvidia (575*) in Fedora 42 and everything’s fine ? That could be a very sweet reason for me to upgrade.
Seriously? It is described in this very thread where the 570 drivers can be downloaded…
You’re talking about the koji links above, right ?
I was talking about the fact that I was expecting for these to be still available from the repos, and not getting into archeological digging to download these one by one - - …
I doubt that this would change anything. The driver and kernel are the same. Is this a laptop with two GPUs?
AFAIK rpmfusion has never offered the previous drivers via dnf downgrade. Dnf downgrade offers the initial drivers when, in your case, F41 was released initially.
1 Like
Well, I actually installed the previous version through dnf. I just told it which version to install and it did it for me.