First time with CUDA -> CUDA driver version is insufficient for CUDA runtime version - Result = FAIL

Accelerated Computing CUDA CUDA Setup and Installation
1

Hi all,

First timer here. Just bought my first “Gaming” type PC, so am totally new to everything Nvidia / GPU. I am familiar with Linux, servers, and open source software in general, just have always made-do with old junk computers.

The GPU is an Nvidia GeForce RTX 3070

I have installed Linux Mint 22 Cinnamon 64 bit.

One application I intend to use is “Flightgear” flight simulator which can make use of CUDA, so have been trying to follow the CUDA Ubuntu network installation guidance here: CUDA Installation Guide for Linux

Which appeared to be successful, until following the post-installation steps I went to test the installation using The Nvidia deviceQuery program from: GitHub - NVIDIA/cuda-samples: Samples for CUDA Developers which demonstrates features in CUDA Toolkit , which returned this result:

$ ./deviceQuery 
./deviceQuery Starting...

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

cudaGetDeviceCount returned 35
-> CUDA driver version is insufficient for CUDA runtime version
Result = FAIL

Info:

$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Tue_Oct_29_23:50:19_PDT_2024
Cuda compilation tools, release 12.6, V12.6.85
Build cuda_12.6.r12.6/compiler.35059454_0

$ lspci
...
08:00.0 VGA compatible controller: NVIDIA Corporation GA104 [GeForce RTX 3070] (rev a1)
08:00.1 Audio device: NVIDIA Corporation GA104 High Definition Audio Controller (rev a1)
...

I then had a stab following this: NVIDIA Transitions Fully Towards Open-Source GPU Kernel Modules | NVIDIA Technical Blog at installing the open source CUDA drivers:

$ sudo apt-get install nvidia-open

This again looked like a promising install, but testing with deviceQuery failed, but with a different code

$ ./deviceQuery 
./deviceQuery Starting...

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

cudaGetDeviceCount returned 100
-> no CUDA-capable device is detected
Result = FAIL

At this point I thought it best to stop randomly installing packages and hoping for the best, as I may make a right mess of my system, and instead I should ask for help/guidance!

Please go gentle on me, ie: if I need to re-install something, please provide a clear guide or link to step by step info as this area is new to me.

Many thanks, Scott.

2

I just had a go with “Nvidia Nsight Compute” that I found in the menu and had this result:

Screenshot from 2024-12-26 12-51-17
Screenshot from 2024-12-26 12-51-171920×1080 191 KB

Don’t know if its helpful…

Cheers, Scott

3

Hang on - after a reboot, I tried again and had this result:

 ./deviceQuery 
./deviceQuery Starting...

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

Detected 1 CUDA Capable device(s)

Device 0: "NVIDIA GeForce RTX 3070"
  CUDA Driver Version / Runtime Version          12.7 / 12.6
  CUDA Capability Major/Minor version number:    8.6
  Total amount of global memory:                 7860 MBytes (8241545216 bytes)
  (046) Multiprocessors, (128) CUDA Cores/MP:    5888 CUDA Cores
  GPU Max Clock rate:                            1725 MHz (1.73 GHz)
  Memory Clock rate:                             7001 Mhz
  Memory Bus Width:                              256-bit
  L2 Cache Size:                                 4194304 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:        102400 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  1536
  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 / 8 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 12.7, CUDA Runtime Version = 12.6, NumDevs = 1
Result = PASS

This seems a lot more positive!