Cuda 11.4.2 docker image driver version mismatch

I’m experiencing issues with the docker image here. As you can see, the sole image being derived from is nvidia/cuda:11.4.1-cudnn8-devel-ubuntu18.04 which is an official nvidia image.

I’ve built the image with the tag rust-cuda and run it with
docker run -it --gpus device=0 --entrypoint /bin/bash rust-cuda

My docker version is Docker version 20.10.7, build 20.10.7-0ubuntu5~20.04.2. The image builds and the container launches without a problem and drops me into its shell.

Building then running the device query sample in this release of the cuda samples Release CUDA Samples v11.4 update 1 · NVIDIA/cuda-samples · GitHub

I get the following output

./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount returned 803
-> system has unsupported display driver / cuda driver combination
Result = FAIL

nvidia-smi returns this from within the container:

+-----------------------------------------------------------------------------+
| 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  On |                  N/A |
|  0%   47C    P8    16W / 230W |    919MiB /  8116MiB |      1%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

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

the host returns

+-----------------------------------------------------------------------------+
| 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  On |                  N/A |
|  0%   60C    P2    40W / 230W |    963MiB /  8116MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1769      G   /usr/lib/xorg/Xorg                168MiB |
|    0   N/A  N/A      6564      G   /usr/lib/xorg/Xorg                610MiB |
|    0   N/A  N/A      7044      G   /opt/piavpn/bin/pia-client          2MiB |
|    0   N/A  N/A      7858      G   /usr/bin/picom                      3MiB |
|    0   N/A  N/A     15126      G   ...AAAAAAAAA= --shared-files      163MiB |
+-----------------------------------------------------------------------------+

as you can see, the driver is 470.86. Checking the tables on this release notes page it looks to me that the driver-cuda version is compatible as for 11.4 update 1 the driver specified for linux x86-64 is >=470.57.02.

Further, I’ve confirmed on the host that this pairing works as the results of deviceQuery are:

./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "NVIDIA GeForce GTX 1070"
  CUDA Driver Version / Runtime Version          11.4 / 11.4
  CUDA Capability Major/Minor version number:    6.1
  Total amount of global memory:                 8116 MBytes (8510701568 bytes)
  (015) Multiprocessors, (128) CUDA Cores/MP:    1920 CUDA Cores
  GPU Max Clock rate:                            1772 MHz (1.77 GHz)
  Memory Clock rate:                             4004 Mhz
  Memory Bus Width:                              256-bit
  L2 Cache Size:                                 2097152 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
  Maximum Layered 1D Texture Size, (num) layers  1D=(32768), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(32768, 32768), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total shared memory per multiprocessor:        98304 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 2 copy engine(s)
  Run time limit on kernels:                     Yes
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device supports Managed Memory:                Yes
  Device supports Compute Preemption:            Yes
  Supports Cooperative Kernel Launch:            Yes
  Supports MultiDevice Co-op Kernel Launch:      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 11.4, CUDA Runtime Version = 11.4, NumDevs = 1
Result = PASS

As it turns out, the issue (still unsure how it was the issue) LD_LIBRARY_PATH being set how it was somehow was causing the issue and revising how I pointed ld to the shared objects by using ldconfig fixed the issue

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