NVIDIA GPU Not Detected on Laptop (Fedora 42) - dmesg Shows "NVRM: No NVIDIA GPU found"

Ask Fedora
, , , ,
1

I’m having trouble getting my NVIDIA GPU detected on my ASUS Vivobook Pro 15 N6506MJ running Fedora Workstation 42 (kernel 6.14.8-300.fc42.x86_64). The system has an Intel Core Ultra 7 155H with an integrated Intel Arc GPU and a discrete NVIDIA GPU RTX 3050. The NVIDIA GPU isn’t detected at all. (Though Dual boot Windows 11 detects it)

I used other available kernel versions including 6.14.6 and 6.14.5 but the issue remains. I am on wayland but testing on xorg didn’t help.

lspci | grep -i nvidia
Shows nothing.

lsmod | grep nvidia
Shows nothing.

lsmod | grep nouveau
Also shows nothing.

nvidia-smi
NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.

sudo dmesg | grep -i nvidia
[ 0.000000] Command line: BOOT_IMAGE=(hd0,gpt6)/vmlinuz-6.14.8-300.fc42.x86_64 root=UUID=a2c5e708-ce54-4888-8fae-ffb131c4a0db ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau modprobe.blacklist=nouveau nvidia-drm.modeset=1 NVreg_PreserveVideoMemoryAllocations=1
[ 0.036396] Kernel command line: BOOT_IMAGE=(hd0,gpt6)/vmlinuz-6.14.8-300.fc42.x86_64 root=UUID=a2c5e708-ce54-4888-8fae-ffb131c4a0db ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau modprobe.blacklist=nouveau nvidia-drm.modeset=1 NVreg_PreserveVideoMemoryAllocations=1
[ 15.220877] nvidia: loading out-of-tree module taints kernel.
[ 15.220884] nvidia: module license ‘NVIDIA’ taints kernel.
[ 15.220887] nvidia: module license taints kernel.
[ 15.359871] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 15.359878] NVRM: No NVIDIA GPU found.
[ 15.361345] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 15.794765] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 15.794773] NVRM: No NVIDIA GPU found.
[ 15.796482] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 133.047708] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 133.047718] NVRM: No NVIDIA GPU found.
[ 133.049263] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 133.506259] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 133.506267] NVRM: No NVIDIA GPU found.
[ 133.508176] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 133.948868] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 133.948876] NVRM: No NVIDIA GPU found.
[ 133.950506] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 146.578531] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 146.578540] NVRM: No NVIDIA GPU found.
[ 146.580250] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 147.032847] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 147.032855] NVRM: No NVIDIA GPU found.
[ 147.034464] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 150.263170] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 150.263177] NVRM: No NVIDIA GPU found.
[ 150.264766] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 150.710735] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 150.710744] NVRM: No NVIDIA GPU found.
[ 150.712528] nvidia-nvlink: Unregistered Nvlink Core, major device number 508
[ 151.146017] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 151.146024] NVRM: No NVIDIA GPU found.
[ 151.147583] nvidia-nvlink: Unregistered Nvlink Core, major device number 508

  • NVIDIA Driver: Installed via akmod-nvidia from RPMFusion.
  • Firmware: linux-firmware and nvidia-gpu-firmware installed.
  • Secure Boot: Enabled, with NVIDIA driver signed via kmodgenca.

modinfo -F version nvidia
570.153.02

My questions are:

  1. Why is the NVIDIA GPU not detected despite nvidia-gpu-firmware being installed, and what does the dmesg error (NVRM: No NVIDIA GPU found) indicate? How to solve the detection issue?
  2. Could TLP or kernel power management (e.g., D3cold) be preventing GPU detection? How can I ensure the GPU is powered? I use TLP.
3

A long shot, and unlikely given that Windows detects it, but perhaps UEFI disables the discrete GPU so the kernel doesn’t see it?

What happens before in the dmesg, e.g. before the nvidia module is initializing?

The lspci output suggests the device isn’t visible at all, so either the kernel tried but failed to register it, or it’s not visible to the kernel (UEFI settings).

If it crashed, it’ll likely have been before timestamp [15…] in dmesg

4

Do you have Fast Boot enabled? I’ve seen some suggestions that in that case, Windows doesn’t properly “release” hardware devices on shutdown.

5

[ 6.674557] iwlwifi 0000:00:14.3: Not valid error log pointer 0x0024B5C0 for RT uCode
…
[ 10.306340] ucsi_acpi USBC000:00: con2: failed to register alt modes
[ 10.400301] ucsi_acpi USBC000:00: error -ETIMEDOUT: PPM init failed
[ 12.684493] wlo1: authenticate with e0:1c:fc:49:55:c2 (local address=f6:0d:aa:d7:64:91)
[ 12.685092] wlo1: send auth to e0:1c:fc:49:55:c2 (try 1/3)
[ 12.712668] wlo1: authenticated
[ 12.713252] wlo1: associate with e0:1c:fc:49:55:c2 (try 1/3)
[ 12.719416] wlo1: RX AssocResp from e0:1c:fc:49:55:c2 (capab=0x411 status=0 aid=1)
[ 12.722935] wlo1: associated
[ 20.808458] Lockdown: systemd-logind: hibernation is restricted; see man kernel_lockdown.7
[ 20.856441] rfkill: input handler disabled
[ 21.542951] Bluetooth: RFCOMM TTY layer initialized
[ 21.542965] Bluetooth: RFCOMM socket layer initialized
[ 21.542967] Bluetooth: RFCOMM ver 1.11
[ 21.732267] nvidia: loading out-of-tree module taints kernel.
[ 21.732274] nvidia: module license ‘NVIDIA’ taints kernel.
[ 21.732274] Disabling lock debugging due to kernel taint
[ 21.732276] nvidia: module license taints kernel.
[ 21.869832] nvidia-nvlink: Nvlink Core is being initialized, major device number 508
[ 21.869839] NVRM: No NVIDIA GPU found.
[ 21.871519] nvidia-nvlink: Unregistered Nvlink Core, major device number 508

These are the dmesg I see that might be related to the issue. Please let me know if more info is needed.

6

No I have disabled fast boot as I remember. I’ll check it again and update this message.

UPDATE: fast boot was enabled in UEFI, probably due to UEFI update, so I disabled it. Still the issue remains.

7

nvidia-gpu-firmware is only useful for nouveau, nvidia doesn’t need or use it.

Is the nvidia gpu enabled in the bios?

8

An LHDB probe of your model shows Nvidia being detected in Fedora 40. As others suggest, the card seems to be disabled at the UEFI level. https://www.nvidia.com/en-us/geforce/forums/geforce-graphics-cards/5/473005/nvidia-rtx-3050-missing-driver-issue-on-a-asus-ga/ looks similar to your issue.

9

Oh OK should I remove it then? I saw in one of the discussions that it worked so I downloaded it.

It sure should be because when I boot to Win 11 it shows Nvidia GPU. But if it needs double checking, please guide me through it.
*UPDATE: I checked UEFI Advanced settings, and the only graphics related setting there was “DVMT Pre-Allocated” which had two options 64M (which was selected) and 32M. I went through other sections related to Intel and most options were enabled, so found nothing related to GPU that can help.

10

My Nvidia card was detected and used to work with the drivers in both fedora 41 and 42. And the link you shared works on a windows device with Asus Armory Crate which my laptop doesn’t have (it’s a vivobook not TUF).

11

I noticed my CPU1 is near 95% all the time (even after reboot)

Used top and here’s the result:

PID USER      PR  NI    VIRT    RES    SHR S  %CPU  %MEM     TIME+ COMMAND                                                                                                                                      
169 root     -51   0       0      0      0 R  69.3   0.0   0:58.41 irq/9-acpi                                                                                                                                   
644 root      20   0       0      0      0 I  22.3   0.0   0:12.45 kworker/0:2-kacpi_notify

I assume acpi is related to my GPU detection issue, right? Anything we can learn from this to resolve the issue?

12

Post the complete output for

sudo dmesg
13

Here it is:

14

The only thing to note is

[    1.727608] pci 0000:00:1c.0: mask Replay Timer Timeout Correctable Errors due to 0000:2c:00.0 hardware defect
1 Like
15

If nouveau works on your Nvidia model you should keep so nouveau can be used when Nvidia fails. Because Nvidia often has issues, I usually just run nouveau for daily use and switch only when I need Nvidia.

UEFI is often similar within different models of a brand but there are changes over time.
Are you using the most recent UEFI updates?

16

Any solutions?
I found these but didn’t worked out (pci=nommconf made the screen go pink on booting):

To address the issue of "Replay Timer Timeout Correctable Errors" due to a hardware defect, you can try adding the **pcie_aspm=off** parameter to the kernel boot options. This parameter disables power management mode, which can sometimes cause such errors.

Additionally, you can add **pci=nommconf** to the kernel boot parameters to disable Memory-Mapped PCI Configuration Space and revert to traditional handling of configuration space.

Unfortunately the Nouveau drivers won’t load as well. lsmod shows no module loading for both nvidia and nouveau. I have to say the reason I needed nvidia was to work on ML projects, so needed cuda.

Yes my UEFI is updated to latest version 311. I assume that might’ve caused the issue in the first place. Should I roll back to older UEFI version?

17

OK weird but out of blue the issue was resolved after moving back and forth between Windows 11 and Fedora Linux. Here’s the dmesg:

sudo dmesg | grep -i nvidia
[sudo] password for parsar2025:
[ 0.000000] Command line: BOOT_IMAGE=(hd0,gpt6)/vmlinuz-6.14.8-300.fc42.x86_64 root=UUID=a2c5e708-ce54-4888-8fae-ffb131c4a0db ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau modprobe.blacklist=nouveau nvidia-drm.modeset=1
[ 0.038708] Kernel command line: BOOT_IMAGE=(hd0,gpt6)/vmlinuz-6.14.8-300.fc42.x86_64 root=UUID=a2c5e708-ce54-4888-8fae-ffb131c4a0db ro rootflags=subvol=root rhgb quiet rd.driver.blacklist=nouveau modprobe.blacklist=nouveau nvidia-drm.modeset=1
[ 4.310048] input: HDA NVidia HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card0/input14
[ 4.310077] input: HDA NVidia HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card0/input15
[ 4.310104] input: HDA NVidia HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card0/input16
[ 4.310121] input: HDA NVidia HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card0/input17
[ 4.353282] nvidia: loading out-of-tree module taints kernel.
[ 4.353289] nvidia: module license ‘NVIDIA’ taints kernel.
[ 4.353291] nvidia: module license taints kernel.
[ 4.492362] nvidia-nvlink: Nvlink Core is being initialized, major device number 509
[ 4.493865] nvidia 0000:01:00.0: enabling device (0000 → 0003)
[ 4.494088] nvidia 0000:01:00.0: vgaarb: VGA decodes changed: olddecodes=io+mem,decodes=none:owns=none
[ 4.539571] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 570.153.02 Tue May 13 16:34:43 UTC 2025
[ 4.579795] nvidia_uvm: module uses symbols nvUvmInterfaceDisableAccessCntr from proprietary module nvidia, inheriting taint.
[ 4.684081] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 570.153.02 Tue May 13 16:21:23 UTC 2025
[ 4.692653] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 6.604305] [drm] Initialized nvidia-drm 0.0.0 for 0000:01:00.0 on minor 0
[ 6.604569] nvidia 0000:01:00.0: [drm] Cannot find any crtc or sizes

What can we find out? Any way to prevent this from happening?

18

In principle Windows shouldn’t be able to write/modify UEFI device settings, but if booting into Windows fixed it perhaps a buggy interaction between Nvidia driver in Windows + UEFI disabled the device or put it in a corrupt state. It’s only a guess from other similar experiences where booting Windows before Linux can cause issues with devices (not helped by soft booting of Windows).
Glad it’s working for you now

1 Like
19

Had same issue with cpu1. I got Asus vivibook pro N6506mv with Ultra 9 H185. Use this to fix ACPI interrupt (works after reboot):

echo “disable” | sudo tee /sys/firmware/acpi/interrupts/gpe6F

Use powertop to to identify your cpu’ s specific GPE interruption. Change ‘gpe6F’ to yours bad GPE (most likely will be defferent.

FYI you can upgrade you ram up to 40GB, not 24GB as Asus says. Done it myself recently and it works.
Good luck

20

Thanks, appreciate all the help from everyone :folded_hands:

21

Wow good point. It is solved now because the gpu is now being detected. But I have it in mind. Thanks :folded_hands:

This is also a good call. Thanks again.