NASP/cuda-samples
85
0
Fork 0
mirror of https://github.com/NVIDIA/cuda-samples.git synced 2026-10-25 22:28:17 +08:00
Code Issues Packages Projects Releases Wiki Activity

CUDA 13.0 removes support for Maxwell, Pascal, and Volta architecture offline compilation

Browse Source
This commit is contained in:
Rob Armstrong
2025-03-27 10:43:00 -07:00
parent eddc6fd7e1
commit 69522dd5b7
187 changed files with 198 additions and 198 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
Samples/7_libNVVM/cuda-c-linking/CMakeLists.txt
View File

@@ -57,12 +57,12 @@ endif()
##############################
### The math library can be built to support a number of GPU platforms using the
### common compute_50 architecture.
### common compute_75 architecture.
### For a collection of specific architecutres, the compiler options
### "-gencode=compute_XX,code=sm_XX"... can be used multiple times in the same
### nvcc invocation.
### The result is bundled into a system library using the 'nvcc -lib' feature.
### >> nvcc -m64 -arch=compute_50 -dc math-funcs.cu -o math-funcs64.o
### >> nvcc -m64 -arch=compute_75 -dc math-funcs.cu -o math-funcs64.o
### >> nvcc -m64 -lib math-funcs64.o -o libmathfuncs64.a
enable_language(CUDA)