Unable to Find v100 GPU in Docker Image

[schrummy14]

Hello, I am working on getting the benchmark to run inside of a docker image with the NVIDIA CUDA runtime (not expecting to use cuda). The image is ubuntu:22.04.

I am receiving the error message that no OpenCL devices are found.

.-----------------------------------------------------------------------------.
|                       ______________   ______________                       |
|                       \   ________  | |  ________   /                       |
|                        \  \       | | | |       /  /                        |
|                         \  \      | | | |      /  /                         |
|                          \  \     | | | |     /  /                          |
|                           \  \_.-"  | |  "-._/  /                           |
|                            \    _.-" _ "-._    /                            |
|                             \.-" _.-" "-._ "-./                             |
|                               .-"  .-"-.  "-.                               |
|                               \  v"     "v  /                               |
|                                \  \     /  /                                |
|                                 \  \   /  /                                 |
|                                  \  \ /  /                                  |
|                                   \  '  /                                   |
|                                    \   /                                    |
|                                     \ /                FluidX3D Version 2.8 |
|                                      '     Copyright (c) Dr. Moritz Lehmann |
|-----------------------------------------------------------------------------|
| Error: There are no OpenCL devices available. Make sure that the OpenCL 1.2 |
|        Runtime for your device is installed. For GPUs it comes by default   |
|        with the graphics driver, for CPUs it has to be installed            |
|        separately.                                                          |
'-----------------------------------------------------------------------------'

I have tried the latest release both pre-compiled and comping inside the docker image. nvidia-smi reports the following:

nimbix@jarvice-job-67537-828ns:~/Documents/FluidX3D-2.8$ nvidia-smi 
Thu Jul  6 20:35:41 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.161.03   Driver Version: 470.161.03   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  Tesla V100-SXM2...  On   | 00000000:00:1E.0 Off |                    0 |
| N/A   37C    P0    25W / 300W |      6MiB / 16160MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

Please let me know if there is any other information that you would like.

[ProjectPhysX]

Hi @schrummy14,

looks like the OpenCL runtime is not installed correctly with the GPU drivers. Same thing happens in Google Colab nowadays, no idea why. If you have sudo permissions, just reinstall the drivers:

!sudo apt update
!sudo apt purge *nvidia* -y
!sudo apt install nvidia-driver-530 -y

Regards, Moritz

[schrummy14]

Thank you for the info. I'll close this out.