RTX 3060 not detected, not even by lspci

Graphics / Linux Linux
,
1

Hello everybody,

I have a MSI GF65 laptop with a RTX 3060 (Max Q, I suppose). I dual boot Ubuntu and Windows and under Windows, everything works great. Under Ubuntu (even after fresh intall), however, the GPU is not detected at all.
The About screen names the graphics device as “Mesa Intel® UHD Graphics (CML GT2)”.
The output of lspci | grep VGA is:
00:02.0 VGA compatible controller: Intel Corporation UHD Graphics (rev 05)
01:00.0 VGA compatible controller: NVIDIA Corporation Device 2520 (rev a1)

I have seen many other posts where the GPU is not detected correctly, but normally at least lspci lists the model name, right? Same for lshw -c video:
*-display
description: VGA compatible controller
product: NVIDIA Corporation
vendor: NVIDIA Corporation
physical id: 0
bus info: pci@0000:01:00.0
logical name: /dev/fb0
version: a1
width: 64 bits
clock: 33MHz
capabilities: pm msi pciexpress vga_controller cap_list fb
configuration: depth=32 latency=0 mode=1920x1080 visual=truecolor xres=1920 yres=1080
resources: iomemory:420-41f iomemory:410-40f memory:4200000000-43ffffffff memory:4100000000-4101ffffff ioport:5000(size=128) memory:b1380000-b13fffff
*-display
description: VGA compatible controller
product: UHD Graphics
vendor: Intel Corporation
physical id: 2
bus info: pci@0000:00:02.0
version: 05
width: 64 bits
clock: 33MHz
capabilities: pciexpress msi pm vga_controller bus_master cap_list
configuration: driver=i915 latency=0
resources: irq:168 memory:b0000000-b0ffffff memory:a0000000-afffffff ioport:4000(size=64) memory:c0000-dffff

Unsurprisingly, nvidia-smi “failed because it couldn’t communicate with the NVIDIA driver.” Also, prime-select nvidia only returns “Info: the nvidia profile is already set”.

I have already tried various suggested solutions from other posts, including removing a file that blacklists nvidiafb, and disabling PSR from grub. Nothing worked so far, but to me it seems like my problem is even more profound than most others I saw here - I’ve never read anything about lspci not being able to name the GPU.

Finally, here’s my nvidia-bug-report.sh output:
nvidia-bug-report.log.gz (447.9 KB)

I would be extremely grateful for any suggestions and hints on how to fix this.
Cheers

1 Like
2

Please set kernel parameter
pci=realloc
to get a proper display name, just update the pci-id database:
sudo update-pciids

1 Like
3

Thank you for your quick reply.
I’ve done as you said, adding pci=realloc to the GRUB_CMDLINE_LINUX_DEFAULT variable in /etc/default/grub. that is enough to add the kernel parameter, right? I am quite a noob on this.

After the second command, the name of the GPU is listed correctly. I didn’t know this could be a simple lookup error, thank you.

However, nvidia-smi still says the same about not being able to communicate with the driver.

4

Did you run
sudo update-grub
and reboot after adding the parameter to actually enable it?
Please create a new nvidia-bug-report.log afterwards.

5

I’m a moron. I definitely rebooted and I thought I had run update-grub, but I just ran it again just in case and now it works. I’m sorry, I must have mixed it up with a previous attempt where I added a line to disable psr.

Thank you for the incredibly fast help and solution.

6

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.