| Table of Contents | ||||||||
|---|---|---|---|---|---|---|---|---|
|
20250129:
The cuda-toolkit/12.8.0 is now available on gcc/14.2.0
Overview
CUDA is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers are able to dramatically speed up computing applications by harnessing the power of GPUs.
...
Use the module name cudato discover versions available and to load the application.
Cuda Versions:
As new versions of cuda are released and compute capabilities of GPUs increase, older cards will stop being supported. As according to: https://forums.developer.nvidia.com/t/nvcc-fatal-unsupported-gpu-architecture-compute-35/247815 “CUDA 12.x has dropped support for Kepler compute 3.x devices. The minimum supported compute capability is 5.0 in CUDA 12.”
For Example: If you trying make-ing some of the NVidia samples, you might see the following error:
| Code Block |
|---|
[@blog1 deviceQuery]$ make
/apps/u/opt/gcc/12.2.0/cuda/12.1.1//bin/nvcc -ccbin g++ -I../../../Common -m64 --threads 0 --std=c++11 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_37,code=sm_37 -gencode arch=compute_50,code=sm_50 -gencode arch=compute_52,code=sm_52 -gencode arch=compute_60,code=sm_60 -gencode arch=compute_61,code=sm_61 -gencode arch=compute_70,code=sm_70 -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -gencode arch=compute_90,code=sm_90 -gencode arch=compute_90,code=compute_90 -o deviceQuery.o -c deviceQuery.cpp
nvcc fatal : Unsupported gpu architecture 'compute_35'
make: *** [Makefile:324: deviceQuery.o] Error 1 |
To resolve, remove the 3x related GPU architecture references from the Makefile:
| Code Block |
|---|
From: SMS ?= 35 37 50 52 60 61 70 75 80 86 90
To : SMS ?= 50 52 60 61 70 75 80 86 90 |