Skip to content

Intel GPU oneDNN upstreaming for library compilation #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
ZhiweiYan-96 wants to merge 6 commits into
base: master
Choose a base branch
from
Open

Intel GPU oneDNN upstreaming for library compilation #2

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
c1082fa
Build oneDNN for xpu as ExternalProject
ZhiweiYan-96 Dec 13, 2023
b146f7d
Rename xpumkldnn project naming
ZhiweiYan-96 Dec 14, 2023
46182f4
format
ZhiweiYan-96 Dec 14, 2023
ff29573
Rename depencies cmake name
ZhiweiYan-96 Dec 21, 2023
c267185
Uinfy cpu&xpu oneDNN cmake
ZhiweiYan-96 Dec 22, 2023
feec482
fix typo
ZhiweiYan-96 Dec 22, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
41 changes: 41 additions & 0 deletions cmake/Modules/FindMKLDNN.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,47 @@ IF(NOT MKLDNN_FOUND)

SET(IDEEP_ROOT "${PROJECT_SOURCE_DIR}/third_party/ideep")
SET(MKLDNN_ROOT "${PROJECT_SOURCE_DIR}/third_party/ideep/mkl-dnn")

if(USE_XPU) # Build oneDNN GPU library
if(CMAKE_CXX_COMPILER_ID STREQUAL "GNU")
set(DNNL_HOST_COMPILER "g++")
# g++ is soft linked to /usr/bin/cxx, oneDNN would not treat it as an absolute path
else()
set(DNNL_HOST_COMPILER DEFAULT)
endif()

set(MAKE_COMMAND "cmake" "--build" ".")
ExternalProject_Add(xpu_mkldnn_proj
SOURCE_DIR ${MKLDNN_ROOT}
BUILD_IN_SOURCE 0
CMAKE_ARGS -DCMAKE_C_COMPILER=icx
-DCMAKE_CXX_COMPILER=icpx
-DCMAKE_CXX_COMPILER_ID=IntelLLVM
-DDNNL_GPU_RUNTIME=SYCL
-DDNNL_CPU_RUNTIME=THREADPOOL
-DDNNL_BUILD_TESTS=OFF
-DDNNL_BUILD_EXAMPLES=OFF
-DDNNL_LIBRARY_TYPE=STATIC
-DDNNL_DPCPP_HOST_COMPILER=${DNNL_HOST_COMPILER} # Use global cxx compiler as host compiler
-G ${CMAKE_GENERATOR} # Align Generator to Torch
BUILD_COMMAND ${MAKE_COMMAND}
INSTALL_COMMAND ""
)

ExternalProject_Get_Property(xpu_mkldnn_proj BINARY_DIR)
set(__XPU__MKLDNN_BUILD_DIR ${BINARY_DIR})
set(XPU_MKLDNN_LIBRARIES ${__XPU_MKLDNN_BUILD_DIR}/src/libdnnl.a)
set(XPU_MKLDNN_INCLUDE ${__XPU_MKLDNN_BUILD_DIR}/src/include)
set(xpu_mkldnn_dep xpu_mkldnn_proj)
# This target would be further linked to libtorch_xpu.so.
# The libtorch_xpu.so would contain Conv&GEMM operators that depend on
# oneDNN primitive implementations inside libdnnl.a.
add_library(xpu_mkldnn INTERFACE)
add_dependencies(xpu_mkldnn xpu_mkldnn_dep)
target_link_libraries(xpu_mkldnn INTERFACE ${__XPU_MKLDNN_BUILD_DIR}/src/libdnnl.a)
target_include_directories(xpu_mkldnn INTERFACE ${XPU_MKLDNN_INCLUDE})
endif()

IF(NOT APPLE AND NOT WIN32 AND NOT BUILD_LITE_INTERPRETER)
MESSAGE("-- Will build oneDNN Graph")
SET(LLGA_ROOT "${PROJECT_SOURCE_DIR}/third_party/ideep/mkl-dnn")
Expand Down