Add option to choose between OMP implementations

CMakeLists.txt

@@ -23,8 +23,7 @@ mxnet_option(USE_CUDA             "Build with CUDA support"   ON)
 mxnet_option(USE_OLDCMAKECUDA     "Build with old cmake cuda" OFF)
 mxnet_option(USE_NCCL             "Use NVidia NCCL with CUDA" OFF)
 mxnet_option(USE_OPENCV           "Build with OpenCV support" ON)
-mxnet_option(USE_OPENMP           "Build with Openmp support" ON)
-mxnet_option(USE_OPENMP_BUNDLED_LLVM "Build with bundled llvm openmp from 3rdparty" OFF)
+mxnet_option(USE_OPENMP           "Build with Openmp support" PLATFORM) # OFF | PLATFORM | BUNDLED
 mxnet_option(USE_CUDNN            "Build with cudnn support"  ON) # one could set CUDNN_ROOT for search path
 mxnet_option(USE_SSE              "Build with x86 SSE instruction support" ON IF NOT ARM)
 mxnet_option(USE_F16C             "Build with x86 F16C instruction support" ON) # autodetects support if ON
@@ -434,11 +433,11 @@ if(USE_OPENMP)
   find_package(OpenMP REQUIRED)
   # This should build on Windows, but there's some problem and I don't have a Windows box, so
   # could a Windows user please fix?
-  if(USE_OPENMP_BUNDLED_LLVM AND EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/3rdparty/openmp/CMakeLists.txt
+  if(USE_OPENMP STREQUAL "BUNDLED" AND EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/3rdparty/openmp/CMakeLists.txt
       AND SYSTEM_ARCHITECTURE STREQUAL "x86_64"
       AND NOT MSVC
       AND NOT CMAKE_CROSSCOMPILING)
-    message("Using bundlded LLVM OpenMP")
+    message("Using bundlded LLVM OpenMP from 3rdparty")
     # Intel/llvm OpenMP: https://github.com/llvm-mirror/openmp
     set(OPENMP_STANDALONE_BUILD TRUE)
     set(LIBOMP_ENABLE_SHARED TRUE)
@@ -451,7 +450,7 @@ if(USE_OPENMP)
     set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} ${OpenMP_CXX_FLAGS}")
     add_definitions(-DMXNET_USE_OPENMP=1)
-  else()
+  elseif(USE_OPENMP STREQUAL "PLATFORM" OR USE_OPENMP STREQUAL "ON")
     message("Using platform provided OpenMP")
     if(OPENMP_FOUND)
       set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${OpenMP_C_FLAGS}")
@@ -460,6 +459,8 @@ if(USE_OPENMP)
       set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} ${OpenMP_EXE_LINKER_FLAGS}")
       add_definitions(-DMXNET_USE_OPENMP=1)
     endif()
+  else()
+    message(FATAL_ERROR "OPENMP_FLAVOR takes values [platform, bundled_llvm]")
   endif()
 elseif(UNIX AND NOT ANDROID)
   list(APPEND mxnet_LINKER_LIBS pthread)

cmake/cmake_options.yml

@@ -20,7 +20,7 @@ USE_CUDA: "OFF" # Build with CUDA support
 USE_OLDCMAKECUDA: "OFF" # Build with old cmake cuda
 USE_NCCL: "OFF" # Use NVidia NCCL with CUDA
 USE_OPENCV: "ON" # Build with OpenCV support
-USE_OPENMP: "ON" # Build with Openmp support
+USE_OPENMP: "PLATFORM" # Build with Openmp support
 USE_CUDNN: "ON" # Build with cudnn support) # one could set CUDNN_ROOT for search path
 USE_SSE: "ON" # Build with x86 SSE instruction support IF NOT ARM
 USE_F16C: "ON" # Build with x86 F16C instruction support) # autodetects support if "ON"