hanchenye-llvm-project/openmp/runtime/cmake/Definitions.cmake

148 lines
5.0 KiB
CMake
Raw Normal View History

#
#//===----------------------------------------------------------------------===//
#//
#// The LLVM Compiler Infrastructure
#//
#// This file is dual licensed under the MIT and the University of Illinois Open
#// Source Licenses. See LICENSE.txt for details.
#//
#//===----------------------------------------------------------------------===//
#
# void append_definitions(string new_flag);
# - appends new_flag to cpp_flags list
macro(append_definitions new_flag)
list(APPEND local_cpp_flags "${new_flag}")
endmacro()
function(append_cpp_flags input_cpp_flags)
set(local_cpp_flags)
append_definitions("-D USE_ITT_BUILD")
append_definitions("-D KMP_ARCH_STR=\"\\\\\"${legal_arch}\\\\\"\"")
append_definitions("-D BUILD_I8")
append_definitions("-D KMP_LIBRARY_FILE=\\\\\"${lib_file}\\\\\"") # yes... you need 5 backslashes...
append_definitions("-D KMP_VERSION_MAJOR=${version}")
# customize to 128 bytes for ppc64
if(${PPC64})
append_definitions("-D CACHE_LINE=128")
else()
append_definitions("-D CACHE_LINE=64")
endif()
append_definitions("-D KMP_ADJUST_BLOCKTIME=1")
append_definitions("-D BUILD_PARALLEL_ORDERED")
append_definitions("-D KMP_ASM_INTRINS")
if(${USE_ITT_NOTIFY})
append_definitions("-D USE_ITT_NOTIFY=1")
else()
append_definitions("-D USE_ITT_NOTIFY=0")
append_definitions("-D INTEL_NO_ITTNOTIFY_API")
endif()
append_definitions("-D INTEL_ITTNOTIFY_PREFIX=__kmp_itt_")
I apologise in advance for the size of this check-in. At Intel we do understand that this is not friendly, and are working to change our internal code-development to make it easier to make development features available more frequently and in finer (more functional) chunks. Unfortunately we haven't got that in place yet, and unpicking this into multiple separate check-ins would be non-trivial, so please bear with me on this one. We should be better in the future. Apologies over, what do we have here? GGC 4.9 compatibility -------------------- * We have implemented the new entrypoints used by code compiled by GCC 4.9 to implement the same functionality in gcc 4.8. Therefore code compiled with gcc 4.9 that used to work will continue to do so. However, there are some other new entrypoints (associated with task cancellation) which are not implemented. Therefore user code compiled by gcc 4.9 that uses these new features will not link against the LLVM runtime. (It remains unclear how to handle those entrypoints, since the GCC interface has potentially unpleasant performance implications for join barriers even when cancellation is not used) --- new parallel entry points --- new entry points that aren't OpenMP 4.0 related These are implemented fully :- GOMP_parallel_loop_dynamic() GOMP_parallel_loop_guided() GOMP_parallel_loop_runtime() GOMP_parallel_loop_static() GOMP_parallel_sections() GOMP_parallel() --- cancellation entry points --- Currently, these only give a runtime error if OMP_CANCELLATION is true because our plain barriers don't check for cancellation while waiting GOMP_barrier_cancel() GOMP_cancel() GOMP_cancellation_point() GOMP_loop_end_cancel() GOMP_sections_end_cancel() --- taskgroup entry points --- These are implemented fully. GOMP_taskgroup_start() GOMP_taskgroup_end() --- target entry points --- These are empty (as they are in libgomp) GOMP_target() GOMP_target_data() GOMP_target_end_data() GOMP_target_update() GOMP_teams() Improvements in Barriers and Fork/Join -------------------------------------- * Barrier and fork/join code is now in its own file (which makes it easier to understand and modify). * Wait/release code is now templated and in its own file; suspend/resume code is also templated * There's a new, hierarchical, barrier, which exploits the cache-hierarchy of the Intel(r) Xeon Phi(tm) coprocessor to improve fork/join and barrier performance. ***BEWARE*** the new source files have *not* been added to the legacy Cmake build system. If you want to use that fixes wil be required. Statistics Collection Code -------------------------- * New code has been added to collect application statistics (if this is enabled at library compile time; by default it is not). The statistics code itself is generally useful, the lightweight timing code uses the X86 rdtsc instruction, so will require changes for other architectures. The intent of this code is not for users to tune their codes but rather 1) For timing code-paths inside the runtime 2) For gathering general properties of OpenMP codes to focus attention on which OpenMP features are most used. Nested Hot Teams ---------------- * The runtime now maintains more state to reduce the overhead of creating and destroying inner parallel teams. This improves the performance of code that repeatedly uses nested parallelism with the same resource allocation. Set the new KMP_HOT_TEAMS_MAX_LEVEL envirable to a depth to enable this (and, of course, OMP_NESTED=true to enable nested parallelism at all). Improved Intel(r) VTune(Tm) Amplifier support --------------------------------------------- * The runtime provides additional information to Vtune via the itt_notify interface to allow it to display better OpenMP specific analyses of load-imbalance. Support for OpenMP Composite Statements --------------------------------------- * Implement new entrypoints required by some of the OpenMP 4.1 composite statements. Improved ifdefs --------------- * More separation of concepts ("Does this platform do X?") from platforms ("Are we compiling for platform Y?"), which should simplify future porting. ScaleMP* contribution --------------------- Stack padding to improve the performance in their environment where cross-node coherency is managed at the page level. Redesign of wait and release code --------------------------------- The code is simplified and performance improved. Bug Fixes --------- *Fixes for Windows multiple processor groups. *Fix Fortran module build on Linux: offload attribute added. *Fix entry names for distribute-parallel-loop construct to be consistent with the compiler codegen. *Fix an inconsistent error message for KMP_PLACE_THREADS environment variable. llvm-svn: 219214
2014-10-08 00:25:50 +08:00
if(${USE_VERSION_SYMBOLS})
append_definitions("-D KMP_USE_VERSION_SYMBOLS")
endif()
#####################
# Windows definitions
if(${WINDOWS})
append_definitions("-D _CRT_SECURE_NO_WARNINGS")
append_definitions("-D _CRT_SECURE_NO_DEPRECATE")
append_definitions("-D _WINDOWS")
append_definitions("-D _WINNT")
append_definitions("-D _WIN32_WINNT=0x0501")
append_definitions("-D KMP_WIN_CDECL")
append_definitions("-D _USRDLL")
if(${DEBUG_BUILD})
append_definitions("-D _ITERATOR_DEBUG_LEVEL=0")
endif()
else() # Other than windows... (Unix based systems, Intel(R) Many Integrated Core Architecture (Intel(R) MIC Architecture), and Mac)
append_definitions("-D _GNU_SOURCE")
append_definitions("-D _REENTRANT")
append_definitions("-D BUILD_TV")
append_definitions("-D USE_CBLKDATA")
if(NOT "${version}" STREQUAL "4")
append_definitions("-D KMP_GOMP_COMPAT")
endif()
endif()
#######################################
# Intel(R) MIC Architecture definitions
if(${MIC})
append_definitions("-D KMP_TDATA_GTID")
else() # Other than Intel(R) MIC Architecture...
append_definitions("-D USE_LOAD_BALANCE")
endif()
##################
# Unix definitions
if(${LINUX})
append_definitions("-D KMP_TDATA_GTID")
endif()
##################################
# Other conditional definitions
append_definitions("-D KMP_USE_ASSERT")
append_definitions("-D GUIDEDLL_EXPORTS")
if(${STUBS_LIBRARY})
append_definitions("-D KMP_STUB")
endif()
if(${DEBUG_BUILD} OR ${RELWITHDEBINFO_BUILD})
append_definitions("-D KMP_DEBUG")
endif()
if(${DEBUG_BUILD})
append_definitions("-D _DEBUG")
append_definitions("-D BUILD_DEBUG")
endif()
if(${STATS_GATHERING})
append_definitions("-D KMP_STATS_ENABLED=1")
else()
append_definitions("-D KMP_STATS_ENABLED=0")
endif()
# OpenMP version flags
set(have_omp_50 0)
set(have_omp_41 0)
set(have_omp_40 0)
set(have_omp_30 0)
if(${omp_version} EQUAL 50 OR ${omp_version} GREATER 50)
set(have_omp_50 1)
endif()
if(${omp_version} EQUAL 41 OR ${omp_version} GREATER 41)
set(have_omp_41 1)
endif()
if(${omp_version} EQUAL 40 OR ${omp_version} GREATER 40)
set(have_omp_40 1)
endif()
if(${omp_version} EQUAL 30 OR ${omp_version} GREATER 30)
set(have_omp_30 1)
endif()
append_definitions("-D OMP_50_ENABLED=${have_omp_50}")
append_definitions("-D OMP_41_ENABLED=${have_omp_41}")
append_definitions("-D OMP_40_ENABLED=${have_omp_40}")
append_definitions("-D OMP_30_ENABLED=${have_omp_30}")
# Architectural definitions
if(${INTEL64} OR ${IA32})
if(${USE_ADAPTIVE_LOCKS})
append_definitions("-D KMP_USE_ADAPTIVE_LOCKS=1")
else()
append_definitions("-D KMP_USE_ADAPTIVE_LOCKS=0")
endif()
append_definitions("-D KMP_DEBUG_ADAPTIVE_LOCKS=0")
else()
append_definitions("-D KMP_USE_ADAPTIVE_LOCKS=0")
append_definitions("-D KMP_DEBUG_ADAPTIVE_LOCKS=0")
endif()
I apologise in advance for the size of this check-in. At Intel we do understand that this is not friendly, and are working to change our internal code-development to make it easier to make development features available more frequently and in finer (more functional) chunks. Unfortunately we haven't got that in place yet, and unpicking this into multiple separate check-ins would be non-trivial, so please bear with me on this one. We should be better in the future. Apologies over, what do we have here? GGC 4.9 compatibility -------------------- * We have implemented the new entrypoints used by code compiled by GCC 4.9 to implement the same functionality in gcc 4.8. Therefore code compiled with gcc 4.9 that used to work will continue to do so. However, there are some other new entrypoints (associated with task cancellation) which are not implemented. Therefore user code compiled by gcc 4.9 that uses these new features will not link against the LLVM runtime. (It remains unclear how to handle those entrypoints, since the GCC interface has potentially unpleasant performance implications for join barriers even when cancellation is not used) --- new parallel entry points --- new entry points that aren't OpenMP 4.0 related These are implemented fully :- GOMP_parallel_loop_dynamic() GOMP_parallel_loop_guided() GOMP_parallel_loop_runtime() GOMP_parallel_loop_static() GOMP_parallel_sections() GOMP_parallel() --- cancellation entry points --- Currently, these only give a runtime error if OMP_CANCELLATION is true because our plain barriers don't check for cancellation while waiting GOMP_barrier_cancel() GOMP_cancel() GOMP_cancellation_point() GOMP_loop_end_cancel() GOMP_sections_end_cancel() --- taskgroup entry points --- These are implemented fully. GOMP_taskgroup_start() GOMP_taskgroup_end() --- target entry points --- These are empty (as they are in libgomp) GOMP_target() GOMP_target_data() GOMP_target_end_data() GOMP_target_update() GOMP_teams() Improvements in Barriers and Fork/Join -------------------------------------- * Barrier and fork/join code is now in its own file (which makes it easier to understand and modify). * Wait/release code is now templated and in its own file; suspend/resume code is also templated * There's a new, hierarchical, barrier, which exploits the cache-hierarchy of the Intel(r) Xeon Phi(tm) coprocessor to improve fork/join and barrier performance. ***BEWARE*** the new source files have *not* been added to the legacy Cmake build system. If you want to use that fixes wil be required. Statistics Collection Code -------------------------- * New code has been added to collect application statistics (if this is enabled at library compile time; by default it is not). The statistics code itself is generally useful, the lightweight timing code uses the X86 rdtsc instruction, so will require changes for other architectures. The intent of this code is not for users to tune their codes but rather 1) For timing code-paths inside the runtime 2) For gathering general properties of OpenMP codes to focus attention on which OpenMP features are most used. Nested Hot Teams ---------------- * The runtime now maintains more state to reduce the overhead of creating and destroying inner parallel teams. This improves the performance of code that repeatedly uses nested parallelism with the same resource allocation. Set the new KMP_HOT_TEAMS_MAX_LEVEL envirable to a depth to enable this (and, of course, OMP_NESTED=true to enable nested parallelism at all). Improved Intel(r) VTune(Tm) Amplifier support --------------------------------------------- * The runtime provides additional information to Vtune via the itt_notify interface to allow it to display better OpenMP specific analyses of load-imbalance. Support for OpenMP Composite Statements --------------------------------------- * Implement new entrypoints required by some of the OpenMP 4.1 composite statements. Improved ifdefs --------------- * More separation of concepts ("Does this platform do X?") from platforms ("Are we compiling for platform Y?"), which should simplify future porting. ScaleMP* contribution --------------------- Stack padding to improve the performance in their environment where cross-node coherency is managed at the page level. Redesign of wait and release code --------------------------------- The code is simplified and performance improved. Bug Fixes --------- *Fixes for Windows multiple processor groups. *Fix Fortran module build on Linux: offload attribute added. *Fix entry names for distribute-parallel-loop construct to be consistent with the compiler codegen. *Fix an inconsistent error message for KMP_PLACE_THREADS environment variable. llvm-svn: 219214
2014-10-08 00:25:50 +08:00
if(${USE_INTERNODE_ALIGNMENT})
append_definitions("-D KMP_USE_INTERNODE_ALIGNMENT=1")
else()
append_definitions("-D KMP_USE_INTERNODE_ALIGNMENT=0")
endif()
set(${input_cpp_flags} "${${input_cpp_flags}}" "${local_cpp_flags}" "${USER_CPP_FLAGS}" "$ENV{CPPFLAGS}" PARENT_SCOPE)
endfunction()