pseXperiments/icicle
1
0
Fork 0
mirror of https://github.com/pseXperiments/icicle.git synced 2026-01-09 13:07:59 -05:00
Code Issues Packages Projects Releases Wiki Activity

Use multi-threaded CUDA compilation to spped up compilation (#439)

Browse Source 
## Describe the changes

Speed up CUDA c++ compile time using multi-threaded compilation
(--split-compile flag).
The tests on 8 core machine show ~2x acceleration.

## Linked Issues

Compiling c++ takes long time
This commit is contained in:
Stas
2024-03-18 14:40:30 -06:00
committed by GitHub
parent 0d806d96ca
commit 0dff1f9302
1 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
icicle/CMakeLists.txt
View File

@@ -60,7 +60,13 @@ else()
endif()
project(icicle LANGUAGES CUDA CXX)
# Check CUDA version and, if possible, enable multi-threaded compilation
if(CMAKE_CUDA_COMPILER_VERSION VERSION_GREATER_EQUAL "12.2")
message(STATUS "Using multi-threaded CUDA compilation.")
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --split-compile 0")
else()
message(STATUS "Can't use multi-threaded CUDA compilation.")
endif()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --expt-relaxed-constexpr")
set(CMAKE_CUDA_FLAGS_RELEASE "")
set(CMAKE_CUDA_FLAGS_DEBUG "${CMAKE_CUDA_FLAGS_DEBUG} -g -G -O0")