dylanyunlong/project_6
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

data: cat SGEMM files from 3 repos into cat_files/

Browse Source 
siboehm/SGEMM_CUDA (19 files):
  siboehm_sgemm.cu, siboehm_runner.cu, siboehm_runner.cuh, siboehm_kernels.cuh
  siboehm_cuBLAS_sgemm.cu, siboehm_simplest_kernel.cu, siboehm_CMakeLists.txt
  siboehm_{1_naive..12_kernel_double_buffering}.cuh

wangzyon/NVIDIA_SGEMM_PRACTICE (12 files):
  wangzyon_sgemm.cu, wangzyon_utils.cu, wangzyon_utils.cuh, wangzyon_kernel.cuh
  wangzyon_CMakeLists.txt, wangzyon_kernel_{1..7}.cuh

edtallison/sgemm-cuda (19 files):
  edtallison_sgemm.cu, edtallison_runner.cu, edtallison_runner.cuh
  edtallison_kernels.cuh, edtallison_cuBLAS_sgemm.cu, edtallison_simplest_kernel.cu
  edtallison_CMakeLists.txt, edtallison_{01_naive..12_kernel_double_buffering}.cuh

cat_files/ total: 25 → 75 files
This commit is contained in:
dylan
2026-08-15 06:59:18 +00:00
parent 284804ac53
commit 7cfa87b5ac
50 changed files with 5612 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
cat_files/siboehm_CMakeLists.txt Normal file
View File

@@ -0,0 +1,36 @@
cmake_minimum_required(VERSION 3.19)
project(NVIDIA_SGEMM_PRACTICE LANGUAGES CXX CUDA)
set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
find_package(CUDA REQUIRED)
# ensure cuda is available
include(CheckLanguage)
check_language(CUDA)
set(CMAKE_CXX_STANDARD 20)
set(CUDA_COMPUTE_CAPABILITY 86)
# in debug mode, add debug symbols to device code
# this disables most optimizations and kills performance
add_compile_options("$<$<AND:$<CONFIG:Debug>,$<COMPILE_LANGUAGE:CUDA>>:-G;-src-in-ptx>")
# add_compile_options("--ptxas-options=-v")
# Configure header file search paths
include_directories(${CUDA_INCLUDE_DIRS})
include_directories(${PROJECT_SOURCE_DIR}/src)
# Configure the source file path to be compiled
aux_source_directory(${PROJECT_SOURCE_DIR}/src SRC)
# generate executable
add_executable(sgemm sgemm.cu ${SRC})
set_target_properties(sgemm PROPERTIES CUDA_ARCHITECTURES ${CUDA_COMPUTE_CAPABILITY})
target_link_libraries(sgemm ${CUDA_LIBRARIES} ${CUDA_CUBLAS_LIBRARIES})
add_executable(cuBLAS_sgemm cuBLAS_sgemm.cu )
set_target_properties(sgemm PROPERTIES CUDA_ARCHITECTURES ${CUDA_COMPUTE_CAPABILITY})
target_link_libraries(cuBLAS_sgemm ${CUDA_LIBRARIES} ${CUDA_CUBLAS_LIBRARIES})
add_executable(simplest_kernel simplest_kernel.cu)
set_target_properties(sgemm PROPERTIES CUDA_ARCHITECTURES ${CUDA_COMPUTE_CAPABILITY})
target_link_libraries(simplest_kernel ${CUDA_LIBRARIES})