commit
048cc30e72
@ -0,0 +1,188 @@
|
||||
if(NOT WITH_GPU)
|
||||
return()
|
||||
endif()
|
||||
|
||||
set(paddle_known_gpu_archs "30 35 50 52 60 61 70")
|
||||
set(paddle_known_gpu_archs7 "30 35 50 52")
|
||||
set(paddle_known_gpu_archs8 "30 35 50 52 60 61")
|
||||
|
||||
######################################################################################
|
||||
# A function for automatic detection of GPUs installed (if autodetection is enabled)
|
||||
# Usage:
|
||||
# detect_installed_gpus(out_variable)
|
||||
function(detect_installed_gpus out_variable)
|
||||
if(NOT CUDA_gpu_detect_output)
|
||||
set(cufile ${PROJECT_BINARY_DIR}/detect_cuda_archs.cu)
|
||||
|
||||
file(WRITE ${cufile} ""
|
||||
"#include <cstdio>\n"
|
||||
"int main() {\n"
|
||||
" int count = 0;\n"
|
||||
" if (cudaSuccess != cudaGetDeviceCount(&count)) return -1;\n"
|
||||
" if (count == 0) return -1;\n"
|
||||
" for (int device = 0; device < count; ++device) {\n"
|
||||
" cudaDeviceProp prop;\n"
|
||||
" if (cudaSuccess == cudaGetDeviceProperties(&prop, device))\n"
|
||||
" std::printf(\"%d.%d \", prop.major, prop.minor);\n"
|
||||
" }\n"
|
||||
" return 0;\n"
|
||||
"}\n")
|
||||
|
||||
execute_process(COMMAND "${CUDA_NVCC_EXECUTABLE}" "-ccbin=${CUDA_HOST_COMPILER}"
|
||||
"--run" "${cufile}"
|
||||
WORKING_DIRECTORY "${PROJECT_BINARY_DIR}/CMakeFiles/"
|
||||
RESULT_VARIABLE nvcc_res OUTPUT_VARIABLE nvcc_out
|
||||
ERROR_QUIET OUTPUT_STRIP_TRAILING_WHITESPACE)
|
||||
|
||||
if(nvcc_res EQUAL 0)
|
||||
# only keep the last line of nvcc_out
|
||||
STRING(REGEX REPLACE ";" "\\\\;" nvcc_out "${nvcc_out}")
|
||||
STRING(REGEX REPLACE "\n" ";" nvcc_out "${nvcc_out}")
|
||||
list(GET nvcc_out -1 nvcc_out)
|
||||
string(REPLACE "2.1" "2.1(2.0)" nvcc_out "${nvcc_out}")
|
||||
set(CUDA_gpu_detect_output ${nvcc_out} CACHE INTERNAL "Returned GPU architetures from detect_installed_gpus tool" FORCE)
|
||||
endif()
|
||||
endif()
|
||||
|
||||
if(NOT CUDA_gpu_detect_output)
|
||||
message(STATUS "Automatic GPU detection failed. Building for all known architectures.")
|
||||
set(${out_variable} ${paddle_known_gpu_archs} PARENT_SCOPE)
|
||||
else()
|
||||
set(${out_variable} ${CUDA_gpu_detect_output} PARENT_SCOPE)
|
||||
endif()
|
||||
endfunction()
|
||||
|
||||
|
||||
########################################################################
|
||||
# Function for selecting GPU arch flags for nvcc based on CUDA_ARCH_NAME
|
||||
# Usage:
|
||||
# select_nvcc_arch_flags(out_variable)
|
||||
function(select_nvcc_arch_flags out_variable)
|
||||
# List of arch names
|
||||
set(archs_names "Kepler" "Maxwell" "Pascal" "All" "Manual")
|
||||
set(archs_name_default "All")
|
||||
if(NOT CMAKE_CROSSCOMPILING)
|
||||
list(APPEND archs_names "Auto")
|
||||
endif()
|
||||
|
||||
# set CUDA_ARCH_NAME strings (so it will be seen as dropbox in CMake-Gui)
|
||||
set(CUDA_ARCH_NAME ${archs_name_default} CACHE STRING "Select target NVIDIA GPU achitecture.")
|
||||
set_property( CACHE CUDA_ARCH_NAME PROPERTY STRINGS "" ${archs_names} )
|
||||
mark_as_advanced(CUDA_ARCH_NAME)
|
||||
|
||||
# verify CUDA_ARCH_NAME value
|
||||
if(NOT ";${archs_names};" MATCHES ";${CUDA_ARCH_NAME};")
|
||||
string(REPLACE ";" ", " archs_names "${archs_names}")
|
||||
message(FATAL_ERROR "Only ${archs_names} architeture names are supported.")
|
||||
endif()
|
||||
|
||||
if(${CUDA_ARCH_NAME} STREQUAL "Manual")
|
||||
set(CUDA_ARCH_BIN ${paddle_known_gpu_archs} CACHE STRING "Specify 'real' GPU architectures to build binaries for, BIN(PTX) format is supported")
|
||||
set(CUDA_ARCH_PTX "50" CACHE STRING "Specify 'virtual' PTX architectures to build PTX intermediate code for")
|
||||
mark_as_advanced(CUDA_ARCH_BIN CUDA_ARCH_PTX)
|
||||
else()
|
||||
unset(CUDA_ARCH_BIN CACHE)
|
||||
unset(CUDA_ARCH_PTX CACHE)
|
||||
endif()
|
||||
|
||||
if(${CUDA_ARCH_NAME} STREQUAL "Kepler")
|
||||
set(cuda_arch_bin "30 35")
|
||||
elseif(${CUDA_ARCH_NAME} STREQUAL "Maxwell")
|
||||
set(cuda_arch_bin "50")
|
||||
elseif(${CUDA_ARCH_NAME} STREQUAL "Pascal")
|
||||
set(cuda_arch_bin "60 61")
|
||||
elseif(${CUDA_ARCH_NAME} STREQUAL "Volta")
|
||||
set(cuda_arch_bin "70")
|
||||
elseif(${CUDA_ARCH_NAME} STREQUAL "All")
|
||||
set(cuda_arch_bin ${paddle_known_gpu_archs})
|
||||
elseif(${CUDA_ARCH_NAME} STREQUAL "Auto")
|
||||
detect_installed_gpus(cuda_arch_bin)
|
||||
else() # (${CUDA_ARCH_NAME} STREQUAL "Manual")
|
||||
set(cuda_arch_bin ${CUDA_ARCH_BIN})
|
||||
endif()
|
||||
|
||||
# remove dots and convert to lists
|
||||
string(REGEX REPLACE "\\." "" cuda_arch_bin "${cuda_arch_bin}")
|
||||
string(REGEX REPLACE "\\." "" cuda_arch_ptx "${CUDA_ARCH_PTX}")
|
||||
string(REGEX MATCHALL "[0-9()]+" cuda_arch_bin "${cuda_arch_bin}")
|
||||
string(REGEX MATCHALL "[0-9]+" cuda_arch_ptx "${cuda_arch_ptx}")
|
||||
list(REMOVE_DUPLICATES cuda_arch_bin)
|
||||
list(REMOVE_DUPLICATES cuda_arch_ptx)
|
||||
|
||||
set(nvcc_flags "")
|
||||
set(nvcc_archs_readable "")
|
||||
|
||||
# Tell NVCC to add binaries for the specified GPUs
|
||||
foreach(arch ${cuda_arch_bin})
|
||||
if(arch MATCHES "([0-9]+)\\(([0-9]+)\\)")
|
||||
# User explicitly specified PTX for the concrete BIN
|
||||
list(APPEND nvcc_flags -gencode arch=compute_${CMAKE_MATCH_2},code=sm_${CMAKE_MATCH_1})
|
||||
list(APPEND nvcc_archs_readable sm_${CMAKE_MATCH_1})
|
||||
else()
|
||||
# User didn't explicitly specify PTX for the concrete BIN, we assume PTX=BIN
|
||||
list(APPEND nvcc_flags -gencode arch=compute_${arch},code=sm_${arch})
|
||||
list(APPEND nvcc_archs_readable sm_${arch})
|
||||
endif()
|
||||
endforeach()
|
||||
|
||||
# Tell NVCC to add PTX intermediate code for the specified architectures
|
||||
foreach(arch ${cuda_arch_ptx})
|
||||
list(APPEND nvcc_flags -gencode arch=compute_${arch},code=compute_${arch})
|
||||
list(APPEND nvcc_archs_readable compute_${arch})
|
||||
endforeach()
|
||||
|
||||
string(REPLACE ";" " " nvcc_archs_readable "${nvcc_archs_readable}")
|
||||
set(${out_variable} ${nvcc_flags} PARENT_SCOPE)
|
||||
set(${out_variable}_readable ${nvcc_archs_readable} PARENT_SCOPE)
|
||||
endfunction()
|
||||
|
||||
message(STATUS "CUDA detected: " ${CUDA_VERSION})
|
||||
if (${CUDA_VERSION} LESS 7.0)
|
||||
set(paddle_known_gpu_archs ${paddle_known_gpu_archs})
|
||||
elseif (${CUDA_VERSION} LESS 8.0) # CUDA 7.x
|
||||
set(paddle_known_gpu_archs ${paddle_known_gpu_archs7})
|
||||
list(APPEND CUDA_NVCC_FLAGS "-D_MWAITXINTRIN_H_INCLUDED")
|
||||
list(APPEND CUDA_NVCC_FLAGS "-D__STRICT_ANSI__")
|
||||
elseif (${CUDA_VERSION} LESS 9.0) # CUDA 8.x
|
||||
set(paddle_known_gpu_archs ${paddle_known_gpu_archs8})
|
||||
list(APPEND CUDA_NVCC_FLAGS "-D_MWAITXINTRIN_H_INCLUDED")
|
||||
list(APPEND CUDA_NVCC_FLAGS "-D__STRICT_ANSI__")
|
||||
# CUDA 8 may complain that sm_20 is no longer supported. Suppress the
|
||||
# warning for now.
|
||||
list(APPEND CUDA_NVCC_FLAGS "-Wno-deprecated-gpu-targets")
|
||||
endif()
|
||||
|
||||
include_directories(${CUDA_INCLUDE_DIRS})
|
||||
list(APPEND EXTERNAL_LIBS ${CUDA_LIBRARIES} ${CUDA_rt_LIBRARY})
|
||||
if(NOT WITH_DSO)
|
||||
list(APPEND EXTERNAL_LIBS ${CUDNN_LIBRARY} ${CUDA_CUBLAS_LIBRARIES} ${CUDA_curand_LIBRARY} ${NCCL_LIBRARY})
|
||||
endif(NOT WITH_DSO)
|
||||
|
||||
# setting nvcc arch flags
|
||||
select_nvcc_arch_flags(NVCC_FLAGS_EXTRA)
|
||||
list(APPEND CUDA_NVCC_FLAGS ${NVCC_FLAGS_EXTRA})
|
||||
message(STATUS "Added CUDA NVCC flags for: ${NVCC_FLAGS_EXTRA_readable}")
|
||||
|
||||
# Set C++11 support
|
||||
set(CUDA_PROPAGATE_HOST_FLAGS OFF)
|
||||
|
||||
# Release/Debug flags set by cmake. Such as -O3 -g -DNDEBUG etc.
|
||||
# So, don't set these flags here.
|
||||
list(APPEND CUDA_NVCC_FLAGS "-std=c++11")
|
||||
list(APPEND CUDA_NVCC_FLAGS "--use_fast_math")
|
||||
list(APPEND CUDA_NVCC_FLAGS "-Xcompiler -fPIC")
|
||||
# Set :expt-relaxed-constexpr to suppress Eigen warnings
|
||||
list(APPEND CUDA_NVCC_FLAGS "--expt-relaxed-constexpr")
|
||||
|
||||
if(CMAKE_BUILD_TYPE STREQUAL "Debug")
|
||||
list(APPEND CUDA_NVCC_FLAGS ${CMAKE_CXX_FLAGS_DEBUG})
|
||||
elseif(CMAKE_BUILD_TYPE STREQUAL "Release")
|
||||
list(APPEND CUDA_NVCC_FLAGS ${CMAKE_CXX_FLAGS_RELEASE})
|
||||
elseif(CMAKE_BUILD_TYPE STREQUAL "RelWithDebInfo")
|
||||
list(APPEND CUDA_NVCC_FLAGS ${CMAKE_CXX_FLAGS_RELWITHDEBINFO})
|
||||
elseif(CMAKE_BUILD_TYPE STREQUAL "MinSizeRel")
|
||||
list(APPEND CUDA_NVCC_FLAGS ${CMAKE_CXX_FLAGS_MINSIZEREL})
|
||||
endif()
|
||||
|
||||
mark_as_advanced(CUDA_BUILD_CUBIN CUDA_BUILD_EMULATION CUDA_VERBOSE_BUILD)
|
||||
mark_as_advanced(CUDA_SDK_ROOT_DIR CUDA_SEPARABLE_COMPILATION)
|
@ -0,0 +1,45 @@
|
||||
# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
#
|
||||
|
||||
IF(MOBILE_INFERENCE)
|
||||
return()
|
||||
ENDIF()
|
||||
|
||||
include (ExternalProject)
|
||||
|
||||
# NOTE: c-ares is needed when linking with grpc.
|
||||
|
||||
SET(CARES_SOURCES_DIR ${THIRD_PARTY_PATH}/cares)
|
||||
SET(CARES_INSTALL_DIR ${THIRD_PARTY_PATH}/install/cares)
|
||||
SET(CARES_INCLUDE_DIR "${CARES_INSTALL_DIR}/include/" CACHE PATH "cares include directory." FORCE)
|
||||
|
||||
ExternalProject_Add(
|
||||
extern_cares
|
||||
GIT_REPOSITORY "https://github.com/c-ares/c-ares.git"
|
||||
GIT_TAG "cares-1_13_0"
|
||||
PREFIX ${CARES_SOURCES_DIR}
|
||||
UPDATE_COMMAND ""
|
||||
CONFIGURE_COMMAND ./buildconf && ./configure --disable-shared --prefix=${CARES_INSTALL_DIR}
|
||||
BUILD_IN_SOURCE 1
|
||||
BUILD_COMMAND make
|
||||
INSTALL_COMMAND make install
|
||||
)
|
||||
|
||||
ADD_LIBRARY(cares STATIC IMPORTED GLOBAL)
|
||||
SET_PROPERTY(TARGET cares PROPERTY IMPORTED_LOCATION
|
||||
"${CARES_INSTALL_DIR}/lib/libcares.a")
|
||||
|
||||
include_directories(${CARES_INCLUDE_DIR})
|
||||
ADD_DEPENDENCIES(cares extern_cares)
|
@ -0,0 +1,66 @@
|
||||
# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
#
|
||||
|
||||
IF(MOBILE_INFERENCE)
|
||||
return()
|
||||
ENDIF()
|
||||
|
||||
include (ExternalProject)
|
||||
|
||||
SET(GRPC_SOURCES_DIR ${THIRD_PARTY_PATH}/grpc)
|
||||
SET(GRPC_INSTALL_DIR ${THIRD_PARTY_PATH}/install/grpc)
|
||||
SET(GRPC_INCLUDE_DIR "${GRPC_INSTALL_DIR}/include/" CACHE PATH "grpc include directory." FORCE)
|
||||
SET(GRPC_CPP_PLUGIN "${GRPC_INSTALL_DIR}/bin/grpc_cpp_plugin" CACHE FILEPATH "GRPC_CPP_PLUGIN" FORCE)
|
||||
IF(APPLE)
|
||||
SET(BUILD_CMD make -n | sed "s/-Werror//g" | sh)
|
||||
ELSE()
|
||||
SET(BUILD_CMD make)
|
||||
ENDIF()
|
||||
|
||||
ExternalProject_Add(
|
||||
extern_grpc
|
||||
DEPENDS protobuf zlib
|
||||
GIT_REPOSITORY "https://github.com/grpc/grpc.git"
|
||||
GIT_TAG "v1.7.x"
|
||||
PREFIX ${GRPC_SOURCES_DIR}
|
||||
UPDATE_COMMAND ""
|
||||
CONFIGURE_COMMAND ""
|
||||
BUILD_IN_SOURCE 1
|
||||
# NOTE(yuyang18):
|
||||
# Disable -Werror, otherwise the compile will fail in MacOS.
|
||||
# It seems that we cannot configure that by make command.
|
||||
# Just dry run make command and remove `-Werror`, then use a shell to run make commands
|
||||
BUILD_COMMAND ${BUILD_CMD} HAS_SYSTEM_PROTOBUF=false -s -j8 static grpc_cpp_plugin
|
||||
INSTALL_COMMAND make prefix=${GRPC_INSTALL_DIR} install
|
||||
)
|
||||
|
||||
# FIXME(typhoonzero): hack to get static lib path, try a better way like merge them.
|
||||
ADD_LIBRARY(grpc++_unsecure STATIC IMPORTED GLOBAL)
|
||||
SET_PROPERTY(TARGET grpc++_unsecure PROPERTY IMPORTED_LOCATION
|
||||
"${GRPC_INSTALL_DIR}/lib/libgrpc++_unsecure.a")
|
||||
|
||||
ADD_LIBRARY(grpc++ STATIC IMPORTED GLOBAL)
|
||||
SET_PROPERTY(TARGET grpc++ PROPERTY IMPORTED_LOCATION
|
||||
"${GRPC_INSTALL_DIR}/lib/libgrpc++.a")
|
||||
ADD_LIBRARY(gpr STATIC IMPORTED GLOBAL)
|
||||
SET_PROPERTY(TARGET gpr PROPERTY IMPORTED_LOCATION
|
||||
"${GRPC_INSTALL_DIR}/lib/libgpr.a")
|
||||
|
||||
ADD_LIBRARY(grpc_unsecure STATIC IMPORTED GLOBAL)
|
||||
SET_PROPERTY(TARGET grpc_unsecure PROPERTY IMPORTED_LOCATION
|
||||
"${GRPC_INSTALL_DIR}/lib/libgrpc_unsecure.a")
|
||||
|
||||
include_directories(${GRPC_INCLUDE_DIR})
|
||||
ADD_DEPENDENCIES(grpc++_unsecure extern_grpc)
|
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in new issue