vLLM
/
vllm
mirror of https://github.com/vllm-project/vllm.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

fix build 

Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
This commit is contained in:
Lucas Wilkinson 2025-06-09 00:38:37 +00:00
parent 5d3b0bc39c
commit d3b51c9bba
2 changed files with 12 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
CMakeLists.txt
View File

@ -436,6 +436,8 @@ if(VLLM_GPU_LANG STREQUAL "CUDA")
MIN_VERSION 12.0 MIN_VERSION 12.0
ARCHS "${MACHETE_ARCHS}" ARCHS "${MACHETE_ARCHS}"
GEN_SCRIPT "${CMAKE_CURRENT_SOURCE_DIR}/csrc/quantization/machete/generate.py" GEN_SCRIPT "${CMAKE_CURRENT_SOURCE_DIR}/csrc/quantization/machete/generate.py"
GEN_PYTHONPATH_PREPEND
"${CMAKE_CURRENT_SOURCE_DIR}/csrc/cutlass_extensions/:${CUTLASS_DIR}/python/"
GEN_GLOB "csrc/quantization/machete/generated/*.cu" GEN_GLOB "csrc/quantization/machete/generated/*.cu"
SRCS "csrc/quantization/machete/machete_pytorch.cu" SRCS "csrc/quantization/machete/machete_pytorch.cu"
VERSION_MSG VERSION_MSG

16
cmake/utils.cmake
View File

@ -40,16 +40,17 @@ function (run_python OUT EXPR ERR_MSG)
endfunction() endfunction()
# Generate CUDA sources via a Python script with caching based on script hash # Generate CUDA sources via a Python script with caching based on script hash
function(generate_cuda_sources NAME SCRIPT GLOB OUT_SRCS) function(generate_cuda_sources NAME SCRIPT PYTHONPATH_PREPEND GLOB OUT_SRCS)
string(TOUPPER "${NAME}" _UPPER_NAME) string(TOUPPER "${NAME}" _UPPER_NAME)
set(_CACHE_VAR "${_UPPER_NAME}_GEN_SCRIPT_HASH") set(_CACHE_VAR "${_UPPER_NAME}_GEN_SCRIPT_HASH")
file(MD5 "${SCRIPT}" _GEN_HASH) file(MD5 "${SCRIPT}" _GEN_HASH)
message(STATUS "${NAME} generation script hash: ${_GEN_HASH}") message(STATUS "${NAME} generation script hash: ${_GEN_HASH}")
message(STATUS "Last run ${NAME} generation script hash: $CACHE{${_CACHE_VAR}}") message(STATUS "Last run ${NAME} generation script hash: $CACHE{${_CACHE_VAR}}")
if(NOT DEFINED CACHE{${_CACHE_VAR}} OR NOT $CACHE{${_CACHE_VAR}} STREQUAL "${_GEN_HASH}") if(NOT DEFINED CACHE{${_CACHE_VAR}} OR NOT $CACHE{${_CACHE_VAR}} STREQUAL "${_GEN_HASH}")
message(STATUS "Running ${NAME} generation script: ${SCRIPT}, with PYTHONPATH prepend: ${PYTHONPATH_PREPEND}")
execute_process( execute_process(
COMMAND ${CMAKE_COMMAND} -E env COMMAND ${CMAKE_COMMAND} -E env
PYTHONPATH=$ENV{PYTHONPATH} PYTHONPATH=${PYTHONPATH_PREPEND}:$ENV{PYTHONPATH}
${Python_EXECUTABLE} "${SCRIPT}" ${Python_EXECUTABLE} "${SCRIPT}"
RESULT_VARIABLE _GEN_RESULT RESULT_VARIABLE _GEN_RESULT
OUTPUT_FILE "${CMAKE_CURRENT_BINARY_DIR}/${NAME}_generation.log" OUTPUT_FILE "${CMAKE_CURRENT_BINARY_DIR}/${NAME}_generation.log"
@ -126,6 +127,7 @@ endfunction()
# [VERSION_MSG <line1> [<line2> ...]] # [VERSION_MSG <line1> [<line2> ...]]
# [NO_ARCH_MSG <line1> [<line2> ...]] # [NO_ARCH_MSG <line1> [<line2> ...]]
# [GEN_SCRIPT <path/to/generate_script.py>] # [GEN_SCRIPT <path/to/generate_script.py>]
# [GEN_PYTHONPATH_PREPEND <str_to_prepend_to_pythonpath>]
# [GEN_GLOB <glob_pattern_for_generated_sources>] # [GEN_GLOB <glob_pattern_for_generated_sources>]
# This will check if `CMAKE_CUDA_COMPILER_VERSION` is greater than or equal # This will check if `CMAKE_CUDA_COMPILER_VERSION` is greater than or equal
# to `MIN_VERSION` and the `cuda_archs_loose_intersection` of `ARCHS` and # to `MIN_VERSION` and the `cuda_archs_loose_intersection` of `ARCHS` and
@ -137,10 +139,11 @@ endfunction()
# 3) append the flags in `FLAGS` to the global `VLLM_GPU_FLAGS` variable. # 3) append the flags in `FLAGS` to the global `VLLM_GPU_FLAGS` variable.
# #
# This will also run GEN_SCRIPT (if supplied and the hash of the script does not # This will also run GEN_SCRIPT (if supplied and the hash of the script does not
# match the latest in the cmake cache), before globbing sources matching # match the latest in the cmake cache), with GEN_PYTHONPATH_PREPEND prepended to
# GEN_GLOB and appending them alongside SRCS (with the gencodes set) # to the PYTHONPATH when calling before globbing sources matching GEN_GLOB
# and appending them alongside SRCS (with the gencodes set)
macro(optional_cuda_sources) macro(optional_cuda_sources)
set(oneValueArgs NAME MIN_VERSION GEN_SCRIPT GEN_GLOB OUT_SRCS_VAR) set(oneValueArgs NAME MIN_VERSION GEN_SCRIPT GEN_PYTHONPATH_PREPEND GEN_GLOB OUT_SRCS_VAR)
set(multiValueArgs ARCHS SRCS FLAGS VERSION_MSG NO_ARCH_MSG) set(multiValueArgs ARCHS SRCS FLAGS VERSION_MSG NO_ARCH_MSG)
cmake_parse_arguments(OCS "" "${oneValueArgs}" "${multiValueArgs}" ${ARGN}) cmake_parse_arguments(OCS "" "${oneValueArgs}" "${multiValueArgs}" ${ARGN})
if(NOT OCS_NAME) if(NOT OCS_NAME)
@ -161,7 +164,8 @@ macro(optional_cuda_sources)
set(_OCS_SRCS ${OCS_SRCS}) set(_OCS_SRCS ${OCS_SRCS})
# Generate sources if a script is provided # Generate sources if a script is provided
if(OCS_GEN_SCRIPT AND OCS_GEN_GLOB) if(OCS_GEN_SCRIPT AND OCS_GEN_GLOB)
generate_cuda_sources(${OCS_NAME} "${OCS_GEN_SCRIPT}" "${OCS_GEN_GLOB}" _OCS_GEN_SRCS) generate_cuda_sources(
${OCS_NAME} "${OCS_GEN_SCRIPT}" "${OCS_GEN_PYTHONPATH_PREPEND}" "${OCS_GEN_GLOB}" _OCS_GEN_SRCS)
list(APPEND _OCS_SRCS ${_OCS_GEN_SRCS}) list(APPEND _OCS_SRCS ${_OCS_GEN_SRCS})
endif() endif()
set_gencode_flags_for_srcs(SRCS "${_OCS_SRCS}" CUDA_ARCHS "${_OCS_ARCHS}") set_gencode_flags_for_srcs(SRCS "${_OCS_SRCS}" CUDA_ARCHS "${_OCS_ARCHS}")