mirror of
https://github.com/deepseek-ai/DeepEP
synced 2025-06-26 18:28:11 +00:00
c8dceba110
* Update CMake files * Use TMA instead of LD/ST for intranode dispatch * Use TMA instead of LD/ST for intranode combine * Adjust configs * Test default configs as well * More warps for combine * Add inter-thread fence * Enable more warps * Do not use TMA for senders * Update configs * Remove useless wait
36 lines
1.4 KiB
CMake
36 lines
1.4 KiB
CMake
# NOTES: this CMake is only for debugging; for setup, please use Torch extension
|
|
cmake_minimum_required(VERSION 3.10)
|
|
project(deep_ep LANGUAGES CUDA CXX)
|
|
set(CMAKE_VERBOSE_MAKEFILE ON)
|
|
|
|
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -O3 -fPIC")
|
|
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -O3 -fPIC")
|
|
set(CUDA_SEPARABLE_COMPILATION ON)
|
|
list(APPEND CUDA_NVCC_FLAGS "-O3")
|
|
list(APPEND CUDA_NVCC_FLAGS "--ptxas-options=--verbose,--register-usage-level=10,--warn-on-local-memory-usage")
|
|
|
|
set(USE_SYSTEM_NVTX on)
|
|
set(CUDA_ARCH_LIST "9.0" CACHE STRING "List of CUDA architectures to compile")
|
|
set(TORCH_CUDA_ARCH_LIST "${CUDA_ARCH_LIST}")
|
|
|
|
find_package(CUDAToolkit REQUIRED)
|
|
find_package(pybind11 REQUIRED)
|
|
find_package(Torch REQUIRED)
|
|
find_package(NVSHMEM REQUIRED HINTS ${NVSHMEM_ROOT_DIR}/lib/cmake/nvshmem)
|
|
|
|
add_library(nvshmem ALIAS nvshmem::nvshmem)
|
|
add_library(nvshmem_host ALIAS nvshmem::nvshmem_host)
|
|
add_library(nvshmem_device ALIAS nvshmem::nvshmem_device)
|
|
|
|
set(CMAKE_CXX_STANDARD 17)
|
|
set(CMAKE_CUDA_STANDARD 17)
|
|
|
|
include_directories(${CUDA_TOOLKIT_ROOT_DIR}/include ${TORCH_INCLUDE_DIRS} ${PYTHON_INCLUDE_DIRS} ${NVSHMEM_INCLUDE_DIR})
|
|
link_directories(${TORCH_INSTALL_PREFIX}/lib ${CUDA_TOOLKIT_ROOT_DIR}/lib ${NVSHMEM_LIB_DIR})
|
|
|
|
add_subdirectory(kernels)
|
|
|
|
# Link CPP and CUDA together
|
|
pybind11_add_module(deep_ep_cpp deep_ep.cpp)
|
|
target_link_libraries(deep_ep_cpp PRIVATE ${EP_CUDA_LIBRARIES} ${TORCH_LIBRARIES} torch_python)
|