So I have been asked to try to speed up some critical code using GPU/CUDA so I pulled out my old laptop to develop & test some CUDA software (In my PC I have and AMD GPU and sadly I currently cannot afford a new GPU for my PC).
The problem is, however, no matter how hard I try to set it all up, it fails to start.
I have been able to reach a stage when I just have to run nvidia-smi or nvidia-settings as a superuser and then it switches to the right GPU and works even without sudo privileges, but I am wondering, how can I fix it so it works right after boot?
I have tried reinstalling drivers, refreshing dkms, reinstalling cuda toolkit, updated kernel to hwe, basically everything that is suggested on the first three pages of google.
To sum it up:
When I write:
nvidia-smi
The result is:
NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
But when I do:
sudo nvidia-smi
nvidia-smi
The output is:
Thu Dec 30 16:36:38 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.86 Driver Version: 470.86 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| 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 ... Off | 00000000:01:00.0 Off | N/A |
| N/A 42C P0 N/A / N/A | 0MiB / 2002MiB | 2% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+
Any way to resolve this?
nvidia-bug-report.log.gz (324.7 KB)