NVIDIA drivers not installed correctly?

Hi.
I have a Lenovo Legion Pro 7 (16ARX8H) with a Ryzen 9 7945HX and an RTX4090.

I have followed this guide to install the NVIDIA drivers. I only went as far as step 7, because I don’t want to have my NVIDIA GPU as the primary GPU.

lspci shows my NVIDIA GPU under the name GN21-X11, which does not seem right to me. neofetch also displays the same weird name.

$ lspci | grep -e VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GN21-X11 (rev a1)
05:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Raphael (rev d8)
$ lspci | grep -e 3D

nvidia-smi however, displays the GPU correctly:

$ nvidia-smi
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 555.58.02              Driver Version: 555.58.02      CUDA Version: 12.5     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce RTX 4090 ...    Off |   00000000:01:00.0 Off |                  N/A |
| N/A   42C    P3             19W /   80W |       2MiB /  16376MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
                                                                                         
+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+

If I try to use the GPU with hashcat, it gives me some errors, but it does correctly recognize the model.

$ hashcat -b
...
Successfully initialized the NVIDIA main driver CUDA runtime library.

Failed to initialize NVIDIA RTC library.

* Device #2: CUDA SDK Toolkit not installed or incorrectly installed.
             CUDA SDK Toolkit required for proper device support and utilization.
             Falling back to OpenCL runtime.

nvmlDeviceGetFanSpeed(): Not Supported

OpenCL API (OpenCL 1.1 Mesa 24.1.5) - Platform #1 [Mesa]
========================================================
* Device #1: AMD Radeon 610M (radeonsi, raphael_mendocino, LLVM 18.1.6, DRM 3.57, 6.10.3-200.fc40.x86_64), skipped

OpenCL API (OpenCL 3.0 CUDA 12.5.85) - Platform #2 [NVIDIA Corporation]
=======================================================================
* Device #2: NVIDIA GeForce RTX 4090 Laptop GPU, 15680/15981 MB (3995 MB allocatable), 76MCU

OpenCL API (OpenCL 3.0 PoCL 5.0  Linux, Release, RELOC, SPIR, LLVM 17.0.6, SLEEF, DISTRO, POCL_DEBUG) - Platform #3 [The pocl project]
======================================================================================================================================
* Device #3: cpu-skylake-avx512-AMD Ryzen 9 7945HX with Radeon Graphics, skipped

I’m using Fedora 40 with KDE.

Did you install the xorg-x11-drv-nvidia-cuda package?
That can be checked with dnf list installed \*nvidia\*.
Please post that output so we can see the result.

The technical codename of RTX 4090 is AD103 GN21-X11. Your NVIDIA GPU is up and running. You can test it with the vkcube command.

hashcat complains about the missing CUDA Toolkit. I don’t know what hashcat is or why it needs CUDA Toolkit but that’s usually for developers. If you need it, you will have to install it from NVIDIA’s website: https://developer.nvidia.com/cuda-downloads.

It looks like you followed the wrong guide. Just to be clear, you followed a guide on how to make the Nvidia GPU your Primary GPU in an X11 environment. . .

Please install the GPU through RPMFusion:
https://rpmfusion.org/Howto/NVIDIA

1 Like

Yes.

Installed Packages
akmod-nvidia.x86_64                               3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
kmod-nvidia-6.10.3-200.fc40.x86_64.x86_64         3:555.58.02-1.fc40          @@commandline                   
kmod-nvidia-6.9.12-200.fc40.x86_64.x86_64         3:555.58.02-1.fc40          @@commandline                   
nvidia-gpu-firmware.noarch                        20240709-1.fc40             @updates                        
nvidia-modprobe.x86_64                            3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
nvidia-persistenced.x86_64                        3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
nvidia-settings.x86_64                            3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia.x86_64                        3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-cuda.x86_64                   3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-cuda-libs.i686                3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-cuda-libs.x86_64              3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-kmodsrc.x86_64                3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-libs.i686                     3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-libs.x86_64                   3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver
xorg-x11-drv-nvidia-power.x86_64                  3:555.58.02-1.fc40          @rpmfusion-nonfree-nvidia-driver

vkcube works. I installed the CUDA toolkit from your link. Although it does say that it’s for Fedora 39, not 40 but after installing it, the warnings disappeared.

Those instructions do exactly as you suggest, but add step 8 to make the nvidia gpu primary.

All those installed packages look good.
It seems vkcube expects something else and you seem to have located the problem.

Hey there ! I only made mention of it because the OP stated . . .

I think they could be tripping over their own feet here.

1 Like