[SOLVED] nvidia module does not load (at startup or at all)

kaizal_btw · 2024-12-12 04:17:26

So, some preamble information, this came up when I was trying to play PoE.
Downloaded from Steam, and doesn't launch.

So I tried running

 vulkaninfo

and got this:

ERROR: [Loader Message] Code 0 : setup_loader_term_phys_devs:  Failed to detect any valid GPUs in the current config
ERROR at /usr/src/debug/vulkan-tools/Vulkan-Tools-1.4.303/vulkaninfo/./vulkaninfo.h:247:vkEnumeratePhysicalDevices failed with ERROR_INITIALIZATION_FAILED

Okay so one of the most basic vulkan functions needed to make anything run doesn't work, maybe it's the nvidia drivers. I have nvidia-open and nvidia-utils installed. So I check if the nvidia kernel modules are running:

However

 lsmod | grep nvidia

returns nothing, so the previous errors make sense now because the required modules are not loaded.

So I try

 sudo modprobe nv

and hit tab to see the options:

nvidia                   nvidia-wmi-ec-backlight  nvme-fc                  nvmem-rave-sp-eeprom     nvmet-fc
nvidia-drm               nvme                     nvme-fcloop              nvmem-rmem               nvmet-rdma
nvidia-modeset           nvme-auth                nvme-keyring             nvme-rdma                nvmet-tcp
nvidia-peermem           nvme-core                nvme-loop                nvmet                    nvsw-sn2201
nvidia-uvm               nvme-fabrics             nvmem                    nvme-tcp                 nv_tco

nvidia is in there so I finish typing and run

 sudo modprobe nvidia

.
I got this output:

 modprobe: ERROR: could not insert 'nvidia': No such device

At this point I really don't know what to do next:

Maybe the output of

 sudo dmesg | grep nvidia

is useful in helping me so I drop it here.

[    4.301809] nvidia: loading out-of-tree module taints kernel.
[    4.301819] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[    4.452440] nvidia-nvlink: Nvlink Core is being initialized, major device number 239
[    4.453212] nvidia 0000:01:00.0: enabling device (0406 -> 0407)
[    4.453358] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=none
               NVRM: nvidia.ko because it does not include the required GPU
               NVRM: [url=http://www.nvidia.com]www.nvidia.com[/url].
[    4.453453] nvidia 0000:01:00.0: probe with driver nvidia failed with error -1
[    4.453632] nvidia-nvlink: Unregistered Nvlink Core, major device number 239
[    4.939385] nvidia-nvlink: Nvlink Core is being initialized, major device number 237
[    4.940378] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=none,decodes=none:owns=none
               NVRM: nvidia.ko because it does not include the required GPU
               NVRM: [url=http://www.nvidia.com]www.nvidia.com[/url].
[    4.940534] nvidia 0000:01:00.0: probe with driver nvidia failed with error -1
[    4.948566] nvidia-nvlink: Unregistered Nvlink Core, major device number 237
[    5.465159] nvidia-nvlink: Nvlink Core is being initialized, major device number 237
[    5.466238] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=none,decodes=none:owns=none
               NVRM: nvidia.ko because it does not include the required GPU
               NVRM: [url=http://www.nvidia.com]www.nvidia.com[/url].

(and the it keeps going on repeat for a while).

I run

 sudo pacman -Syu

before doing this steps and I am up to date. Also am on vanilla arch btw.

Last edited by kaizal_btw (2024-12-13 02:20:28)

pvtvega · 2024-12-12 04:35:42

NVRM: nvidia.ko because it does not include the required GPU

This indicates that the Nvidia driver you have installed does not recognize this GPU. What GPU do you have? Are you sure you installed the correct driver for your video card?

kaizal_btw · 2024-12-12 05:53:21

pvtvega wrote:

NVRM: nvidia.ko because it does not include the required GPU
This indicates that the Nvidia driver you have installed does not recognize this GPU. What GPU do you have? Are you sure you installed the correct driver for your video card?

fastfetch output:

 
                `ooo/                    OS: Arch Linux x86_64
               `+oooo:                   Host: Predator G3-571 (V1.22)
              `+oooooo:                  Kernel: Linux 6.12.4-arch1-1
              -+oooooo+:                 Uptime: 2 hours, 37 mins
            `/:-:++oooo+:                Packages: 839 (pacman)
           `/++++/+++++++:               Shell: bash 5.2.37
          `/++++++++++++++:              Display (eDP-1): 1920x1080 @ 60 Hz in 16" [Built-in]
         `/+++ooooooooooooo/`            WM: i3 (X11)
        ./ooosssso++osssssso+`           Cursor: Adwaita
       .oossssso-````/ossssss+`          Terminal: kitty 0.37.0
      -osssssso.      :ssssssso.         Terminal Font: DejaVuSansMono (18pt)
     :osssssss/        osssso+++.        CPU: Intel(R) Core(TM) i7-7700HQ (8) @ 3.80 GHz
    /ossssssss/        +ssssooo/-        GPU 1: NVIDIA GeForce GTX 1060 Mobile [Discrete]
  `/ossssso+/:-        -:/+osssso+-      GPU 2: Intel HD Graphics 630 @ 1.10 GHz [Integrated]
 `+sso+:-`                 `.-/+oso:     Memory: 2.63 GiB / 15.50 GiB (17%)
`++:.                           `-/+/    Swap: Disabled
.`                                 `/    Disk (/): 113.61 GiB / 232.69 GiB (49%) - ext4
                                         Disk (/home/Juan-Pedro/data_mount_point): 373.58 GiB / 686.60 GiB (54%) - ext4
                                         Local IP (wlp2s0): 192.168.1.6/24
                                         Battery (0x4B384234314341): 100% [AC Connected]
                                         Locale: en_US.UTF-8

seth · 2024-12-12 11:07:32

That's a pascal chip, are you trying to use nvidia-open? (doesn't support pascal but only turing or newer)
Also please don't grep dmesg for random tokens - nvrm is typically more important than nvidia, you're dropping have of the module message and there could also be bus errors leading up to this.
If it's not because of nvidia-open, post the entire dmesg.

kaizal_btw · 2024-12-13 02:19:50

seth wrote:

That's a pascal chip, are you trying to use nvidia-open? (doesn't support pascal but only turing or newer)

You were indeed right, thanks, getting rid of nvidia-open and installing nvidia worked (vulkaninfo and vkcube don't shit the bed).

TL;DR for future readers: Don't be dumb like me and install nvidia-open on old cards.

pvtvega · 2024-12-13 13:23:27

kaizal_btw wrote:

TL;DR for future readers: Don't be dumb like me and install nvidia-open on old cards.

As always future readers, please check the wiki. Section 1 here shows which nvidia drivers are required for which cards: https://wiki.archlinux.org/title/NVIDIA

Arch Linux

#1 2024-12-12 04:17:26

[SOLVED] nvidia module does not load (at startup or at all)

#2 2024-12-12 04:35:42

Re: [SOLVED] nvidia module does not load (at startup or at all)

#3 2024-12-12 05:53:21

Re: [SOLVED] nvidia module does not load (at startup or at all)

#4 2024-12-12 11:07:32

Re: [SOLVED] nvidia module does not load (at startup or at all)

#5 2024-12-13 02:19:50

Re: [SOLVED] nvidia module does not load (at startup or at all)

#6 2024-12-13 13:23:27

Re: [SOLVED] nvidia module does not load (at startup or at all)

Board footer