Nvidia-smi showing 'no devices were found' after installing driver from system76

After installing nvidia driver from system76 using below commands I’m seeing no devices found for nvidia-smi
I need to connect an external monitor - please help me out in installing correct driver

attaching bug report for your reference

exec bash 
sudo apt purge ~nnvidia 
sudo apt clean 
sudo apt update -m 
sudo dpkg --configure -a 
sudo apt install -f 
sudo apt full-upgrade 
sudo apt autoremove 
sudo apt install system76-driver-nvidia 
sudo shutdown -r now
(base) rahilshiraz@pop-os:~$ lshw -c video
WARNING: you should run this program as super-user.
       description: VGA compatible controller
       product: TU117M
       vendor: NVIDIA Corporation
       physical id: 0
       bus info: pci@0000:01:00.0
       version: a1
       width: 64 bits
       clock: 33MHz
       capabilities: vga_controller cap_list rom
       configuration: driver=nvidia latency=0
       resources: irq:16 memory:c6000000-c6ffffff memory:b0000000-bfffffff memory:c0000000-c1ffffff ioport:5000(size=128) memory:c5000000-c507ffff
       description: VGA compatible controller
       product: CometLake-H GT2 [UHD Graphics]
       vendor: Intel Corporation
       physical id: 2
       bus info: pci@0000:00:02.0
       logical name: /dev/fb0
       version: 05
       width: 64 bits
       clock: 33MHz
       capabilities: vga_controller bus_master cap_list rom fb
       configuration: depth=32 driver=i915 latency=0 resolution=1920,1080
       resources: irq:140 memory:c3000000-c3ffffff memory:a0000000-afffffff ioport:6000(size=64) memory:c0000-dffff

nvidia-bug-report.log.gz (197.8 KB)

Hi there @shirazrahil63, welcoem to the NVIDIA developer forums.

Is it possible that you still have conflicting kernel modules installed?

Oct 15 10:52:09 pop-os kernel: [    4.802813] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  470.199.02  Thu May 11 11:46:56 UTC 2023


Oct 15 11:06:40 pop-os kernel: [    4.663333] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  535.113.01  Tue Sep 12 19:41:24 UTC 2023

Maybe sudo apt purge ~nnvidia does not do what you intended? Did you mean something like this: sudo apt-get remove --purge '^nvidia-.*'?

reinstalled after running this sudo apt-get remove --purge '^nvidia-.*'

I see this now when i run nvidia-smi after choosing Hybrid Graphics on Pop OS

(base) rahilshiraz@pop-os:~$ nvidia-smi
Unable to determine the device handle for GPU0000:01:00.0: Unknown Error

my external monitor is turned on but is freezed and unable to move my cursor over there
If I choose NVIDIA graphics both laptop and monitor is freezed.

Please help me out to connect my external monitor

nvidia-bug-report.log.gz (209.3 KB)

Did you reboot after doing the purge AND after reinstallation? Did you reinstall in console mode with all NVIDIA kernel modules unloaded? Because the kernel.log still shows both driver versions.

In any case, there might actually be an issue with your BIOS or even the PCI or GPU Hardware.

[    5.900389] NVRM: Xid (PCI:0000:01:00): 79, pid='<unknown>', name=<unknown>, GPU has fallen off the bus.

This can mean a variety of things, but quite often it has to do with either the PCIe BUS or Temperature/Power issues of the GPU. BUS problems can be resolved by re-seating the GPU, but with a Laptop that obviously is not an option. Power should also not be a problem, but you might want to check the temperatures of the system and GPU in particular.

There is also this:

Oct 15 10:50:31 pop-os kernel: [   32.035024] NVRM: Xid (PCI:0000:01:00): 61, pid=621, 0d20(31fc) 00000000 00000000

Which indicates an internal Microcontroller issue. And

[    0.467145] pci 0000:01:00.0: can't claim BAR 6 [mem 0xfff80000-0xffffffff pref]: no compatible bridge window

which might or might not cause issues with memory allocation for GPU BAR mappings. You might want to check if you have some resizable BAR settings in BIOS which are enabled by accident. Disable it if it is enabled, since GTX1650 mobile does not support ReBAR.

Beyond that, I don’t really have more ideas.