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