Ubuntu 22.04 3090 and CUDA - what am I doing wrong?

I must be doing something wrong, not sure what it is:

New ubuntu minimal install. Which works.

When I install Driver Version: 520.61.05 CUDA Version: 11.8 from here
the machine complains about an obsolete keyring, but otherwise seems to be
build everything as it should.

But I don’t get any graphics output.

dmesg shows:

[ 11.819457] nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing.

nvidia-smi takes forever,

nvidia-smi
Fri Oct 28 19:22:46 2022

±----------------------------------------------------------------------------+
| NVIDIA-SMI 520.61.05 Driver Version: 520.61.05 CUDA Version: 11.8 |
|-------------------------------±---------------------±---------------------+
| 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 … On | 00000000:01:00.0 Off | N/A |
| 53% 50C P0 131W / 350W | 43MiB / 24576MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 931 G /usr/lib/xorg/Xorg 37MiB |
| 0 N/A N/A 1846 G /usr/bin/gnome-shell 4MiB |
±----------------------------------------------------------------------------+

dmesg | grep nvidia
[ 6.915674] audit: type=1400 audit(1666977225.124:6): apparmor=“STATUS” operation=“profile_load” profile=“unconfined” name=“nvidia_modprobe” pid=609 comm=“apparmor_parser”
[ 6.915677] audit: type=1400 audit(1666977225.124:7): apparmor=“STATUS” operation=“profile_load” profile=“unconfined” name=“nvidia_modprobe//kmod” pid=609 comm=“apparmor_parser”
[ 7.043542] nvidia: loading out-of-tree module taints kernel.
[ 7.043549] nvidia: module license ‘NVIDIA’ taints kernel.
[ 7.053855] nvidia: module verification failed: signature and/or required key missing - tainting kernel
[ 7.061064] nvidia-nvlink: Nvlink Core is being initialized, major device number 510
[ 7.061900] nvidia 0000:01:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=none
[ 7.243651] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 520.61.05 Thu Sep 29 05:29:37 UTC 2022
[ 7.247147] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 7.247148] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
[ 7.267212] nvidia_uvm: module uses symbols from proprietary module nvidia, inheriting taint.
[ 7.268979] nvidia-uvm: Loaded the UVM driver, major device number 508.
[ 11.819457] nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing.
[ 223.893819] nvidia-modeset: WARNING: GPU:0: Unable to read EDID for display device LG Electronics LG HDR 4K (HDMI-1)
[ 512.919976] nvidia-modeset: ERROR: GPU:0: Error while waiting for GPU progress: 0x0000c67e:0 3:0:216:208

Here the bug report.

nvidia-bug-report.log.gz (364.9 KB)

The hardware is happy with ubuntu 22.04 and 5.10 - just that I need CUDA …

Drivers 515.76 and 520.65.01 have a known bug with hdmi connections. Please try downgrading to 520.56 or use a DP connection.