enne2/qwentts.cpp
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

cmake: emit Pascal (61-real) only on CUDA 12.x toolkits

Browse Source 
CUDA 13 removed offline compilation for pre-Turing architectures,
so keeping 61-real in the default list makes nvcc fail with
"Unsupported gpu architecture compute_61" on 13+ toolkits.
Split the version check: 13+ drops 61-real, 12.8+ keeps it with
Blackwell, older 12.x keeps it without.
This commit is contained in:
Pascal
2026-08-07 08:04:06 +02:00
parent 978fc08f5a
commit 07f837494a
1 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CMakeLists.txt
+6 -2
View File
@@ -49,11 +49,15 @@ endif()
# CUDA architectures: cover Pascal to Blackwell for distributed binaries.
# Pascal (61-real) is SASS-only, no virtual/PTX entry: it's the oldest
# supported card and doesn't need to seed forward JIT compat for anything
# older, unlike the 75-virtual baseline which does that for 7.5+.
# older, unlike the 75-virtual baseline which does that for 7.5+. CUDA 13
# removed offline compilation for pre-Turing architectures, so 61-real is
# only emitted on 12.x toolkits; Blackwell (120a/121a) needs 12.8+.
# Users can override with -DCMAKE_CUDA_ARCHITECTURES=native for local builds.
if(NOT DEFINED CMAKE_CUDA_ARCHITECTURES)
find_package(CUDAToolkit QUIET)
if(CUDAToolkit_FOUND AND CUDAToolkit_VERSION VERSION_GREATER_EQUAL "12.8")
if(CUDAToolkit_FOUND AND CUDAToolkit_VERSION VERSION_GREATER_EQUAL "13.0")
set(CMAKE_CUDA_ARCHITECTURES "75-virtual;80-virtual;86-real;89-real;120a-real;121a-real")
elseif(CUDAToolkit_FOUND AND CUDAToolkit_VERSION VERSION_GREATER_EQUAL "12.8")
set(CMAKE_CUDA_ARCHITECTURES "61-real;75-virtual;80-virtual;86-real;89-real;120a-real;121a-real")
else()
set(CMAKE_CUDA_ARCHITECTURES "61-real;75-virtual;80-virtual;86-real;89-real")