Skip to content

GitLab

"...git@developer.sourcefind.cn:renzhc/diffusers_dcu.git" did not exist on "36f43ea75ab7cdf9b04f72bced0b1ab22036c21c"
Commit fac152d2 authored by muozturk's avatar muozturk
Browse files

fp32 and fp64 are done

parent 89f5d30a
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 3151 additions and 1632 deletions
example/64_complex_contraction_bilinear/CMakeCache.txt 0 → 100644
View file @ fac152d2
# This is the CMakeCache file.
# For build in directory: /data/ck/composable_kernel/example/64_complex_contraction
# It was generated by CMake: /usr/bin/cmake
# You can edit this file to change values found and used by cmake.
# If you do not want to change any of the values, simply exit the editor.
# If you do want to change a value, simply edit, save, and exit the editor.
# The syntax for the file is as follows:
# KEY:TYPE=VALUE
# KEY is the name of a variable in the cache.
# TYPE is a hint to GUIs for the type of VALUE, DO NOT EDIT TYPE!.
# VALUE is the current value for the KEY.
########################
# EXTERNAL cache entries
########################
//Path to a program.
CMAKE_ADDR2LINE:FILEPATH=/usr/bin/addr2line
//Path to a program.
CMAKE_AR:FILEPATH=/usr/bin/ar
//For backwards compatibility, what version of CMake commands and
// syntax should this version of CMake try to support.
CMAKE_BACKWARDS_COMPATIBILITY:STRING=2.4
//Choose the type of build, options are: None Debug Release RelWithDebInfo
// MinSizeRel ...
CMAKE_BUILD_TYPE:STRING=
//Enable/Disable color output during build.
CMAKE_COLOR_MAKEFILE:BOOL=ON
//CXX compiler
CMAKE_CXX_COMPILER:FILEPATH=/usr/bin/c++
//A wrapper around 'ar' adding the appropriate '--plugin' option
// for the GCC compiler
CMAKE_CXX_COMPILER_AR:FILEPATH=/usr/bin/gcc-ar-9
//A wrapper around 'ranlib' adding the appropriate '--plugin' option
// for the GCC compiler
CMAKE_CXX_COMPILER_RANLIB:FILEPATH=/usr/bin/gcc-ranlib-9
//Flags used by the CXX compiler during all build types.
CMAKE_CXX_FLAGS:STRING=
//Flags used by the CXX compiler during DEBUG builds.
CMAKE_CXX_FLAGS_DEBUG:STRING=-g
//Flags used by the CXX compiler during MINSIZEREL builds.
CMAKE_CXX_FLAGS_MINSIZEREL:STRING=-Os -DNDEBUG
//Flags used by the CXX compiler during RELEASE builds.
CMAKE_CXX_FLAGS_RELEASE:STRING=-O3 -DNDEBUG
//Flags used by the CXX compiler during RELWITHDEBINFO builds.
CMAKE_CXX_FLAGS_RELWITHDEBINFO:STRING=-O2 -g -DNDEBUG
//C compiler
CMAKE_C_COMPILER:FILEPATH=/usr/bin/cc
//A wrapper around 'ar' adding the appropriate '--plugin' option
// for the GCC compiler
CMAKE_C_COMPILER_AR:FILEPATH=/usr/bin/gcc-ar-9
//A wrapper around 'ranlib' adding the appropriate '--plugin' option
// for the GCC compiler
CMAKE_C_COMPILER_RANLIB:FILEPATH=/usr/bin/gcc-ranlib-9
//Flags used by the C compiler during all build types.
CMAKE_C_FLAGS:STRING=
//Flags used by the C compiler during DEBUG builds.
CMAKE_C_FLAGS_DEBUG:STRING=-g
//Flags used by the C compiler during MINSIZEREL builds.
CMAKE_C_FLAGS_MINSIZEREL:STRING=-Os -DNDEBUG
//Flags used by the C compiler during RELEASE builds.
CMAKE_C_FLAGS_RELEASE:STRING=-O3 -DNDEBUG
//Flags used by the C compiler during RELWITHDEBINFO builds.
CMAKE_C_FLAGS_RELWITHDEBINFO:STRING=-O2 -g -DNDEBUG
//Path to a program.
CMAKE_DLLTOOL:FILEPATH=CMAKE_DLLTOOL-NOTFOUND
//Flags used by the linker during all build types.
CMAKE_EXE_LINKER_FLAGS:STRING=
//Flags used by the linker during DEBUG builds.
CMAKE_EXE_LINKER_FLAGS_DEBUG:STRING=
//Flags used by the linker during MINSIZEREL builds.
CMAKE_EXE_LINKER_FLAGS_MINSIZEREL:STRING=
//Flags used by the linker during RELEASE builds.
CMAKE_EXE_LINKER_FLAGS_RELEASE:STRING=
//Flags used by the linker during RELWITHDEBINFO builds.
CMAKE_EXE_LINKER_FLAGS_RELWITHDEBINFO:STRING=
//Enable/Disable output of compile commands during generation.
CMAKE_EXPORT_COMPILE_COMMANDS:BOOL=OFF
//Install path prefix, prepended onto install directories.
CMAKE_INSTALL_PREFIX:PATH=/usr/local
//Path to a program.
CMAKE_LINKER:FILEPATH=/usr/bin/ld
//Path to a program.
CMAKE_MAKE_PROGRAM:FILEPATH=/usr/bin/make
//Flags used by the linker during the creation of modules during
// all build types.
CMAKE_MODULE_LINKER_FLAGS:STRING=
//Flags used by the linker during the creation of modules during
// DEBUG builds.
CMAKE_MODULE_LINKER_FLAGS_DEBUG:STRING=
//Flags used by the linker during the creation of modules during
// MINSIZEREL builds.
CMAKE_MODULE_LINKER_FLAGS_MINSIZEREL:STRING=
//Flags used by the linker during the creation of modules during
// RELEASE builds.
CMAKE_MODULE_LINKER_FLAGS_RELEASE:STRING=
//Flags used by the linker during the creation of modules during
// RELWITHDEBINFO builds.
CMAKE_MODULE_LINKER_FLAGS_RELWITHDEBINFO:STRING=
//Path to a program.
CMAKE_NM:FILEPATH=/usr/bin/nm
//Path to a program.
CMAKE_OBJCOPY:FILEPATH=/usr/bin/objcopy
//Path to a program.
CMAKE_OBJDUMP:FILEPATH=/usr/bin/objdump
//Value Computed by CMake
CMAKE_PROJECT_DESCRIPTION:STATIC=
//Value Computed by CMake
CMAKE_PROJECT_HOMEPAGE_URL:STATIC=
//Value Computed by CMake
CMAKE_PROJECT_NAME:STATIC=Project
//Path to a program.
CMAKE_RANLIB:FILEPATH=/usr/bin/ranlib
//Path to a program.
CMAKE_READELF:FILEPATH=/usr/bin/readelf
//Flags used by the linker during the creation of shared libraries
// during all build types.
CMAKE_SHARED_LINKER_FLAGS:STRING=
//Flags used by the linker during the creation of shared libraries
// during DEBUG builds.
CMAKE_SHARED_LINKER_FLAGS_DEBUG:STRING=
//Flags used by the linker during the creation of shared libraries
// during MINSIZEREL builds.
CMAKE_SHARED_LINKER_FLAGS_MINSIZEREL:STRING=
//Flags used by the linker during the creation of shared libraries
// during RELEASE builds.
CMAKE_SHARED_LINKER_FLAGS_RELEASE:STRING=
//Flags used by the linker during the creation of shared libraries
// during RELWITHDEBINFO builds.
CMAKE_SHARED_LINKER_FLAGS_RELWITHDEBINFO:STRING=
//If set, runtime paths are not added when installing shared libraries,
// but are added when building.
CMAKE_SKIP_INSTALL_RPATH:BOOL=NO
//If set, runtime paths are not added when using shared libraries.
CMAKE_SKIP_RPATH:BOOL=NO
//Flags used by the linker during the creation of static libraries
// during all build types.
CMAKE_STATIC_LINKER_FLAGS:STRING=
//Flags used by the linker during the creation of static libraries
// during DEBUG builds.
CMAKE_STATIC_LINKER_FLAGS_DEBUG:STRING=
//Flags used by the linker during the creation of static libraries
// during MINSIZEREL builds.
CMAKE_STATIC_LINKER_FLAGS_MINSIZEREL:STRING=
//Flags used by the linker during the creation of static libraries
// during RELEASE builds.
CMAKE_STATIC_LINKER_FLAGS_RELEASE:STRING=
//Flags used by the linker during the creation of static libraries
// during RELWITHDEBINFO builds.
CMAKE_STATIC_LINKER_FLAGS_RELWITHDEBINFO:STRING=
//Path to a program.
CMAKE_STRIP:FILEPATH=/usr/bin/strip
//If this value is on, makefiles will be generated without the
// .SILENT directive, and all commands will be echoed to the console
// during the make. This is useful for debugging only. With Visual
// Studio IDE projects all commands are done without /nologo.
CMAKE_VERBOSE_MAKEFILE:BOOL=FALSE
//Single output directory for building all executables.
EXECUTABLE_OUTPUT_PATH:PATH=
//Single output directory for building all libraries.
LIBRARY_OUTPUT_PATH:PATH=
//Value Computed by CMake
Project_BINARY_DIR:STATIC=/data/ck/composable_kernel/example/64_complex_contraction
//Value Computed by CMake
Project_SOURCE_DIR:STATIC=/data/ck/composable_kernel/example/64_complex_contraction
########################
# INTERNAL cache entries
########################
//ADVANCED property for variable: CMAKE_ADDR2LINE
CMAKE_ADDR2LINE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_AR
CMAKE_AR-ADVANCED:INTERNAL=1
//This is the directory where this CMakeCache.txt was created
CMAKE_CACHEFILE_DIR:INTERNAL=/data/ck/composable_kernel/example/64_complex_contraction
//Major version of cmake used to create the current loaded cache
CMAKE_CACHE_MAJOR_VERSION:INTERNAL=3
//Minor version of cmake used to create the current loaded cache
CMAKE_CACHE_MINOR_VERSION:INTERNAL=16
//Patch version of cmake used to create the current loaded cache
CMAKE_CACHE_PATCH_VERSION:INTERNAL=3
//ADVANCED property for variable: CMAKE_COLOR_MAKEFILE
CMAKE_COLOR_MAKEFILE-ADVANCED:INTERNAL=1
//Path to CMake executable.
CMAKE_COMMAND:INTERNAL=/usr/bin/cmake
//Path to cpack program executable.
CMAKE_CPACK_COMMAND:INTERNAL=/usr/bin/cpack
//Path to ctest program executable.
CMAKE_CTEST_COMMAND:INTERNAL=/usr/bin/ctest
//ADVANCED property for variable: CMAKE_CXX_COMPILER
CMAKE_CXX_COMPILER-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_COMPILER_AR
CMAKE_CXX_COMPILER_AR-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_COMPILER_RANLIB
CMAKE_CXX_COMPILER_RANLIB-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_FLAGS
CMAKE_CXX_FLAGS-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_FLAGS_DEBUG
CMAKE_CXX_FLAGS_DEBUG-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_FLAGS_MINSIZEREL
CMAKE_CXX_FLAGS_MINSIZEREL-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_FLAGS_RELEASE
CMAKE_CXX_FLAGS_RELEASE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_CXX_FLAGS_RELWITHDEBINFO
CMAKE_CXX_FLAGS_RELWITHDEBINFO-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_COMPILER
CMAKE_C_COMPILER-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_COMPILER_AR
CMAKE_C_COMPILER_AR-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_COMPILER_RANLIB
CMAKE_C_COMPILER_RANLIB-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_FLAGS
CMAKE_C_FLAGS-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_FLAGS_DEBUG
CMAKE_C_FLAGS_DEBUG-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_FLAGS_MINSIZEREL
CMAKE_C_FLAGS_MINSIZEREL-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_FLAGS_RELEASE
CMAKE_C_FLAGS_RELEASE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_C_FLAGS_RELWITHDEBINFO
CMAKE_C_FLAGS_RELWITHDEBINFO-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_DLLTOOL
CMAKE_DLLTOOL-ADVANCED:INTERNAL=1
//Executable file format
CMAKE_EXECUTABLE_FORMAT:INTERNAL=ELF
//ADVANCED property for variable: CMAKE_EXE_LINKER_FLAGS
CMAKE_EXE_LINKER_FLAGS-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_EXE_LINKER_FLAGS_DEBUG
CMAKE_EXE_LINKER_FLAGS_DEBUG-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_EXE_LINKER_FLAGS_MINSIZEREL
CMAKE_EXE_LINKER_FLAGS_MINSIZEREL-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_EXE_LINKER_FLAGS_RELEASE
CMAKE_EXE_LINKER_FLAGS_RELEASE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_EXE_LINKER_FLAGS_RELWITHDEBINFO
CMAKE_EXE_LINKER_FLAGS_RELWITHDEBINFO-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_EXPORT_COMPILE_COMMANDS
CMAKE_EXPORT_COMPILE_COMMANDS-ADVANCED:INTERNAL=1
//Name of external makefile project generator.
CMAKE_EXTRA_GENERATOR:INTERNAL=
//Name of generator.
CMAKE_GENERATOR:INTERNAL=Unix Makefiles
//Generator instance identifier.
CMAKE_GENERATOR_INSTANCE:INTERNAL=
//Name of generator platform.
CMAKE_GENERATOR_PLATFORM:INTERNAL=
//Name of generator toolset.
CMAKE_GENERATOR_TOOLSET:INTERNAL=
//Source directory with the top level CMakeLists.txt file for this
// project
CMAKE_HOME_DIRECTORY:INTERNAL=/data/ck/composable_kernel/example/64_complex_contraction
//Install .so files without execute permission.
CMAKE_INSTALL_SO_NO_EXE:INTERNAL=1
//ADVANCED property for variable: CMAKE_LINKER
CMAKE_LINKER-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_MAKE_PROGRAM
CMAKE_MAKE_PROGRAM-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_MODULE_LINKER_FLAGS
CMAKE_MODULE_LINKER_FLAGS-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_MODULE_LINKER_FLAGS_DEBUG
CMAKE_MODULE_LINKER_FLAGS_DEBUG-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_MODULE_LINKER_FLAGS_MINSIZEREL
CMAKE_MODULE_LINKER_FLAGS_MINSIZEREL-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_MODULE_LINKER_FLAGS_RELEASE
CMAKE_MODULE_LINKER_FLAGS_RELEASE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_MODULE_LINKER_FLAGS_RELWITHDEBINFO
CMAKE_MODULE_LINKER_FLAGS_RELWITHDEBINFO-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_NM
CMAKE_NM-ADVANCED:INTERNAL=1
//number of local generators
CMAKE_NUMBER_OF_MAKEFILES:INTERNAL=1
//ADVANCED property for variable: CMAKE_OBJCOPY
CMAKE_OBJCOPY-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_OBJDUMP
CMAKE_OBJDUMP-ADVANCED:INTERNAL=1
//Platform information initialized
CMAKE_PLATFORM_INFO_INITIALIZED:INTERNAL=1
//ADVANCED property for variable: CMAKE_RANLIB
CMAKE_RANLIB-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_READELF
CMAKE_READELF-ADVANCED:INTERNAL=1
//Path to CMake installation.
CMAKE_ROOT:INTERNAL=/usr/share/cmake-3.16
//ADVANCED property for variable: CMAKE_SHARED_LINKER_FLAGS
CMAKE_SHARED_LINKER_FLAGS-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_SHARED_LINKER_FLAGS_DEBUG
CMAKE_SHARED_LINKER_FLAGS_DEBUG-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_SHARED_LINKER_FLAGS_MINSIZEREL
CMAKE_SHARED_LINKER_FLAGS_MINSIZEREL-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_SHARED_LINKER_FLAGS_RELEASE
CMAKE_SHARED_LINKER_FLAGS_RELEASE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_SHARED_LINKER_FLAGS_RELWITHDEBINFO
CMAKE_SHARED_LINKER_FLAGS_RELWITHDEBINFO-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_SKIP_INSTALL_RPATH
CMAKE_SKIP_INSTALL_RPATH-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_SKIP_RPATH
CMAKE_SKIP_RPATH-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_STATIC_LINKER_FLAGS
CMAKE_STATIC_LINKER_FLAGS-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_STATIC_LINKER_FLAGS_DEBUG
CMAKE_STATIC_LINKER_FLAGS_DEBUG-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_STATIC_LINKER_FLAGS_MINSIZEREL
CMAKE_STATIC_LINKER_FLAGS_MINSIZEREL-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_STATIC_LINKER_FLAGS_RELEASE
CMAKE_STATIC_LINKER_FLAGS_RELEASE-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_STATIC_LINKER_FLAGS_RELWITHDEBINFO
CMAKE_STATIC_LINKER_FLAGS_RELWITHDEBINFO-ADVANCED:INTERNAL=1
//ADVANCED property for variable: CMAKE_STRIP
CMAKE_STRIP-ADVANCED:INTERNAL=1
//uname command
CMAKE_UNAME:INTERNAL=/usr/bin/uname
//ADVANCED property for variable: CMAKE_VERBOSE_MAKEFILE
CMAKE_VERBOSE_MAKEFILE-ADVANCED:INTERNAL=1
example/64_complex_contraction_bilinear/CMakeFiles/3.16.3/CMakeCCompiler.cmake 0 → 100644
View file @ fac152d2
set(CMAKE_C_COMPILER "/usr/bin/cc")
set(CMAKE_C_COMPILER_ARG1 "")
set(CMAKE_C_COMPILER_ID "GNU")
set(CMAKE_C_COMPILER_VERSION "9.4.0")
set(CMAKE_C_COMPILER_VERSION_INTERNAL "")
set(CMAKE_C_COMPILER_WRAPPER "")
set(CMAKE_C_STANDARD_COMPUTED_DEFAULT "11")
set(CMAKE_C_COMPILE_FEATURES "c_std_90;c_function_prototypes;c_std_99;c_restrict;c_variadic_macros;c_std_11;c_static_assert")
set(CMAKE_C90_COMPILE_FEATURES "c_std_90;c_function_prototypes")
set(CMAKE_C99_COMPILE_FEATURES "c_std_99;c_restrict;c_variadic_macros")
set(CMAKE_C11_COMPILE_FEATURES "c_std_11;c_static_assert")
set(CMAKE_C_PLATFORM_ID "Linux")
set(CMAKE_C_SIMULATE_ID "")
set(CMAKE_C_COMPILER_FRONTEND_VARIANT "")
set(CMAKE_C_SIMULATE_VERSION "")
set(CMAKE_AR "/usr/bin/ar")
set(CMAKE_C_COMPILER_AR "/usr/bin/gcc-ar-9")
set(CMAKE_RANLIB "/usr/bin/ranlib")
set(CMAKE_C_COMPILER_RANLIB "/usr/bin/gcc-ranlib-9")
set(CMAKE_LINKER "/usr/bin/ld")
set(CMAKE_MT "")
set(CMAKE_COMPILER_IS_GNUCC 1)
set(CMAKE_C_COMPILER_LOADED 1)
set(CMAKE_C_COMPILER_WORKS TRUE)
set(CMAKE_C_ABI_COMPILED TRUE)
set(CMAKE_COMPILER_IS_MINGW )
set(CMAKE_COMPILER_IS_CYGWIN )
if(CMAKE_COMPILER_IS_CYGWIN)
set(CYGWIN 1)
set(UNIX 1)
endif()
set(CMAKE_C_COMPILER_ENV_VAR "CC")
if(CMAKE_COMPILER_IS_MINGW)
set(MINGW 1)
endif()
set(CMAKE_C_COMPILER_ID_RUN 1)
set(CMAKE_C_SOURCE_FILE_EXTENSIONS c;m)
set(CMAKE_C_IGNORE_EXTENSIONS h;H;o;O;obj;OBJ;def;DEF;rc;RC)
set(CMAKE_C_LINKER_PREFERENCE 10)
# Save compiler ABI information.
set(CMAKE_C_SIZEOF_DATA_PTR "8")
set(CMAKE_C_COMPILER_ABI "ELF")
set(CMAKE_C_LIBRARY_ARCHITECTURE "x86_64-linux-gnu")
if(CMAKE_C_SIZEOF_DATA_PTR)
set(CMAKE_SIZEOF_VOID_P "${CMAKE_C_SIZEOF_DATA_PTR}")
endif()
if(CMAKE_C_COMPILER_ABI)
set(CMAKE_INTERNAL_PLATFORM_ABI "${CMAKE_C_COMPILER_ABI}")
endif()
if(CMAKE_C_LIBRARY_ARCHITECTURE)
set(CMAKE_LIBRARY_ARCHITECTURE "x86_64-linux-gnu")
endif()
set(CMAKE_C_CL_SHOWINCLUDES_PREFIX "")
if(CMAKE_C_CL_SHOWINCLUDES_PREFIX)
set(CMAKE_CL_SHOWINCLUDES_PREFIX "${CMAKE_C_CL_SHOWINCLUDES_PREFIX}")
endif()
set(CMAKE_C_IMPLICIT_INCLUDE_DIRECTORIES "/usr/lib/gcc/x86_64-linux-gnu/9/include;/usr/local/include;/usr/include/x86_64-linux-gnu;/usr/include")
set(CMAKE_C_IMPLICIT_LINK_LIBRARIES "gcc;gcc_s;c;gcc;gcc_s")
set(CMAKE_C_IMPLICIT_LINK_DIRECTORIES "/usr/lib/gcc/x86_64-linux-gnu/9;/usr/lib/x86_64-linux-gnu;/usr/lib;/lib/x86_64-linux-gnu;/lib")
set(CMAKE_C_IMPLICIT_LINK_FRAMEWORK_DIRECTORIES "")
example/64_complex_contraction_bilinear/CMakeFiles/3.16.3/CMakeCXXCompiler.cmake 0 → 100644
View file @ fac152d2
set(CMAKE_CXX_COMPILER "/usr/bin/c++")
set(CMAKE_CXX_COMPILER_ARG1 "")
set(CMAKE_CXX_COMPILER_ID "GNU")
set(CMAKE_CXX_COMPILER_VERSION "9.4.0")
set(CMAKE_CXX_COMPILER_VERSION_INTERNAL "")
set(CMAKE_CXX_COMPILER_WRAPPER "")
set(CMAKE_CXX_STANDARD_COMPUTED_DEFAULT "14")
set(CMAKE_CXX_COMPILE_FEATURES "cxx_std_98;cxx_template_template_parameters;cxx_std_11;cxx_alias_templates;cxx_alignas;cxx_alignof;cxx_attributes;cxx_auto_type;cxx_constexpr;cxx_decltype;cxx_decltype_incomplete_return_types;cxx_default_function_template_args;cxx_defaulted_functions;cxx_defaulted_move_initializers;cxx_delegating_constructors;cxx_deleted_functions;cxx_enum_forward_declarations;cxx_explicit_conversions;cxx_extended_friend_declarations;cxx_extern_templates;cxx_final;cxx_func_identifier;cxx_generalized_initializers;cxx_inheriting_constructors;cxx_inline_namespaces;cxx_lambdas;cxx_local_type_template_args;cxx_long_long_type;cxx_noexcept;cxx_nonstatic_member_init;cxx_nullptr;cxx_override;cxx_range_for;cxx_raw_string_literals;cxx_reference_qualified_functions;cxx_right_angle_brackets;cxx_rvalue_references;cxx_sizeof_member;cxx_static_assert;cxx_strong_enums;cxx_thread_local;cxx_trailing_return_types;cxx_unicode_literals;cxx_uniform_initialization;cxx_unrestricted_unions;cxx_user_literals;cxx_variadic_macros;cxx_variadic_templates;cxx_std_14;cxx_aggregate_default_initializers;cxx_attribute_deprecated;cxx_binary_literals;cxx_contextual_conversions;cxx_decltype_auto;cxx_digit_separators;cxx_generic_lambdas;cxx_lambda_init_captures;cxx_relaxed_constexpr;cxx_return_type_deduction;cxx_variable_templates;cxx_std_17;cxx_std_20")
set(CMAKE_CXX98_COMPILE_FEATURES "cxx_std_98;cxx_template_template_parameters")
set(CMAKE_CXX11_COMPILE_FEATURES "cxx_std_11;cxx_alias_templates;cxx_alignas;cxx_alignof;cxx_attributes;cxx_auto_type;cxx_constexpr;cxx_decltype;cxx_decltype_incomplete_return_types;cxx_default_function_template_args;cxx_defaulted_functions;cxx_defaulted_move_initializers;cxx_delegating_constructors;cxx_deleted_functions;cxx_enum_forward_declarations;cxx_explicit_conversions;cxx_extended_friend_declarations;cxx_extern_templates;cxx_final;cxx_func_identifier;cxx_generalized_initializers;cxx_inheriting_constructors;cxx_inline_namespaces;cxx_lambdas;cxx_local_type_template_args;cxx_long_long_type;cxx_noexcept;cxx_nonstatic_member_init;cxx_nullptr;cxx_override;cxx_range_for;cxx_raw_string_literals;cxx_reference_qualified_functions;cxx_right_angle_brackets;cxx_rvalue_references;cxx_sizeof_member;cxx_static_assert;cxx_strong_enums;cxx_thread_local;cxx_trailing_return_types;cxx_unicode_literals;cxx_uniform_initialization;cxx_unrestricted_unions;cxx_user_literals;cxx_variadic_macros;cxx_variadic_templates")
set(CMAKE_CXX14_COMPILE_FEATURES "cxx_std_14;cxx_aggregate_default_initializers;cxx_attribute_deprecated;cxx_binary_literals;cxx_contextual_conversions;cxx_decltype_auto;cxx_digit_separators;cxx_generic_lambdas;cxx_lambda_init_captures;cxx_relaxed_constexpr;cxx_return_type_deduction;cxx_variable_templates")
set(CMAKE_CXX17_COMPILE_FEATURES "cxx_std_17")
set(CMAKE_CXX20_COMPILE_FEATURES "cxx_std_20")
set(CMAKE_CXX_PLATFORM_ID "Linux")
set(CMAKE_CXX_SIMULATE_ID "")
set(CMAKE_CXX_COMPILER_FRONTEND_VARIANT "")
set(CMAKE_CXX_SIMULATE_VERSION "")
set(CMAKE_AR "/usr/bin/ar")
set(CMAKE_CXX_COMPILER_AR "/usr/bin/gcc-ar-9")
set(CMAKE_RANLIB "/usr/bin/ranlib")
set(CMAKE_CXX_COMPILER_RANLIB "/usr/bin/gcc-ranlib-9")
set(CMAKE_LINKER "/usr/bin/ld")
set(CMAKE_MT "")
set(CMAKE_COMPILER_IS_GNUCXX 1)
set(CMAKE_CXX_COMPILER_LOADED 1)
set(CMAKE_CXX_COMPILER_WORKS TRUE)
set(CMAKE_CXX_ABI_COMPILED TRUE)
set(CMAKE_COMPILER_IS_MINGW )
set(CMAKE_COMPILER_IS_CYGWIN )
if(CMAKE_COMPILER_IS_CYGWIN)
set(CYGWIN 1)
set(UNIX 1)
endif()
set(CMAKE_CXX_COMPILER_ENV_VAR "CXX")
if(CMAKE_COMPILER_IS_MINGW)
set(MINGW 1)
endif()
set(CMAKE_CXX_COMPILER_ID_RUN 1)
set(CMAKE_CXX_SOURCE_FILE_EXTENSIONS C;M;c++;cc;cpp;cxx;m;mm;CPP)
set(CMAKE_CXX_IGNORE_EXTENSIONS inl;h;hpp;HPP;H;o;O;obj;OBJ;def;DEF;rc;RC)
foreach (lang C OBJC OBJCXX)
if (CMAKE_${lang}_COMPILER_ID_RUN)
foreach(extension IN LISTS CMAKE_${lang}_SOURCE_FILE_EXTENSIONS)
list(REMOVE_ITEM CMAKE_CXX_SOURCE_FILE_EXTENSIONS ${extension})
endforeach()
endif()
endforeach()
set(CMAKE_CXX_LINKER_PREFERENCE 30)
set(CMAKE_CXX_LINKER_PREFERENCE_PROPAGATES 1)
# Save compiler ABI information.
set(CMAKE_CXX_SIZEOF_DATA_PTR "8")
set(CMAKE_CXX_COMPILER_ABI "ELF")
set(CMAKE_CXX_LIBRARY_ARCHITECTURE "x86_64-linux-gnu")
if(CMAKE_CXX_SIZEOF_DATA_PTR)
set(CMAKE_SIZEOF_VOID_P "${CMAKE_CXX_SIZEOF_DATA_PTR}")
endif()
if(CMAKE_CXX_COMPILER_ABI)
set(CMAKE_INTERNAL_PLATFORM_ABI "${CMAKE_CXX_COMPILER_ABI}")
endif()
if(CMAKE_CXX_LIBRARY_ARCHITECTURE)
set(CMAKE_LIBRARY_ARCHITECTURE "x86_64-linux-gnu")
endif()
set(CMAKE_CXX_CL_SHOWINCLUDES_PREFIX "")
if(CMAKE_CXX_CL_SHOWINCLUDES_PREFIX)
set(CMAKE_CL_SHOWINCLUDES_PREFIX "${CMAKE_CXX_CL_SHOWINCLUDES_PREFIX}")
endif()
set(CMAKE_CXX_IMPLICIT_INCLUDE_DIRECTORIES "/usr/include/c++/9;/usr/include/x86_64-linux-gnu/c++/9;/usr/include/c++/9/backward;/usr/lib/gcc/x86_64-linux-gnu/9/include;/usr/local/include;/usr/include/x86_64-linux-gnu;/usr/include")
set(CMAKE_CXX_IMPLICIT_LINK_LIBRARIES "stdc++;m;gcc_s;gcc;c;gcc_s;gcc")
set(CMAKE_CXX_IMPLICIT_LINK_DIRECTORIES "/usr/lib/gcc/x86_64-linux-gnu/9;/usr/lib/x86_64-linux-gnu;/usr/lib;/lib/x86_64-linux-gnu;/lib")
set(CMAKE_CXX_IMPLICIT_LINK_FRAMEWORK_DIRECTORIES "")
example/64_complex_contraction_bilinear/CMakeFiles/3.16.3/CMakeSystem.cmake 0 → 100644
View file @ fac152d2
set(CMAKE_HOST_SYSTEM "Linux-5.15.0-56-generic")
set(CMAKE_HOST_SYSTEM_NAME "Linux")
set(CMAKE_HOST_SYSTEM_VERSION "5.15.0-56-generic")
set(CMAKE_HOST_SYSTEM_PROCESSOR "x86_64")
set(CMAKE_SYSTEM "Linux-5.15.0-56-generic")
set(CMAKE_SYSTEM_NAME "Linux")
set(CMAKE_SYSTEM_VERSION "5.15.0-56-generic")
set(CMAKE_SYSTEM_PROCESSOR "x86_64")
set(CMAKE_CROSSCOMPILING "FALSE")
set(CMAKE_SYSTEM_LOADED 1)
example/64_complex_contraction_bilinear/CMakeFiles/cmake.check_cache 0 → 100644
View file @ fac152d2
# This file is generated by cmake for dependency checking of the CMakeCache.txt file
example/64_complex_contraction_bilinear/CMakeLists.txt 0 → 100755
View file @ fac152d2
add_example_executable(example_complex_contraction_bilinear_xdl_fp32 complex_contraction_bilinear_xdl_fp32.cpp)
add_example_executable(example_complex_contraction_scale_xdl_fp32 complex_contraction_scale_xdl_fp32.cpp)
add_example_executable(example_complex_contraction_bilinear_xdl_fp64 complex_contraction_bilinear_xdl_fp64.cpp)
add_example_executable(example_complex_contraction_scale_xdl_fp64 complex_contraction_scale_xdl_fp64.cpp)
example/64_complex_contraction_bilinear/common_instances.hpp 0 → 100644
View file @ fac152d2
This diff is collapsed.
example/64_complex_contraction_bilinear/complex_contraction_bilinear_xdl_fp32.cpp 0 → 100755
View file @ fac152d2
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/ck.hpp"
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
#include "common_instances.hpp"
using ADataType = F32;
using BDataType = F32;
using AccDataType = F32;
using CShuffleDataType = F32;
using DDataType = F32;
using DsDataType = ck::Tuple<DDataType>;
using EDataType = F32;
using ComputeDataType = F32;
static constexpr ck::index_t NumDimM = 2;
static constexpr ck::index_t NumDimN = 2;
static constexpr ck::index_t NumDimK = 2;
using AElementOp = ck::tensor_operation::element_wise::PassThrough;
using BElementOp = ck::tensor_operation::element_wise::PassThrough;
using CDEElementOp = ck::tensor_operation::element_wise::Bilinear;
using DeviceOpInstanceKKNN = DeviceOpInstanceKK_Generic<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstanceKNNN = DeviceOpInstanceKN_Generic<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstanceMKNN = DeviceOpInstanceMK_Generic<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstanceMNNN = DeviceOpInstanceMN_Generic<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstance = DeviceOpInstanceKKNN;
#include "run_complex_contraction_bilinear_example.inc"
int main(int argc, char* argv[]) { return run_complex_contraction_bilinear_example(argc, argv); }
example/64_complex_contraction_bilinear/complex_contraction_bilinear_xdl_fp64.cpp 0 → 100755
View file @ fac152d2
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/ck.hpp"
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
#include "common_instances.hpp"
using ADataType = F64;
using BDataType = F64;
using AccDataType = F64;
using CShuffleDataType = F64;
using DDataType = F64;
using DsDataType = ck::Tuple<DDataType>;
using EDataType = F64;
using ComputeDataType = F64;
static constexpr ck::index_t NumDimM = 2;
static constexpr ck::index_t NumDimN = 2;
static constexpr ck::index_t NumDimK = 2;
using AElementOp = ck::tensor_operation::element_wise::PassThrough;
using BElementOp = ck::tensor_operation::element_wise::PassThrough;
using CDEElementOp = ck::tensor_operation::element_wise::Bilinear;
using DeviceOpInstanceKKNN = DeviceOpInstanceKK_FP64<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstanceKNNN = DeviceOpInstanceKN_FP64<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstanceMKNN = DeviceOpInstanceMK_FP64<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstanceMNNN = DeviceOpInstanceMN_FP64<NumDimM,
NumDimN,
NumDimK,
ADataType,
BDataType,
AccDataType,
CShuffleDataType,
DsDataType,
EDataType,
ComputeDataType,
AElementOp,
BElementOp,
CDEElementOp>;
using DeviceOpInstance = DeviceOpInstanceKKNN;
#include "run_complex_contraction_bilinear_example.inc"
int main(int argc, char* argv[]) { return run_complex_contraction_bilinear_example(argc, argv); }
example/complex_contraction/4D_kernel.hpp deleted 100644 → 0
View file @ 89f5d30a
This diff is collapsed.
example/complex_contraction/Makefile deleted 100644 → 0
View file @ 89f5d30a
4D_kernel:
hipcc -O3 --offload-arch=gfx90a main.cpp -o $@
clean:
rm 4D_kernel
\ No newline at end of file
example/complex_contraction/main.cpp deleted 100644 → 0
View file @ 89f5d30a
//
// Sample Code:
//
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "4D_kernel.hpp"
// #define DEBUG_CORRECTNESS
//#define DEBUG_SIMPLE_CORRECTNESS
void pre_Initializing_Input_Tensors();
void post_Correctness();
// Initialize t3 (t3_temp), 9 t2 and 9 v2.
void pre_Initializing_Input_Tensors(Complex* h_C, Complex* h_C_chk, int size_C, Complex* h_A, int size_A, Complex* h_B, int size_B)
{
// t3
int i, j;
for (i = 0; i < size_C; i++)
{
h_C[i].re = 0.0;
h_C_chk[i].re = 0.0;
h_C[i].im = 0.0;
h_C_chk[i].im = 0.0;
}
for (j = 0; j < size_A; j++)
{
h_A[j].re = ((float)rand() / RAND_MAX);
h_A[j].im = ((float)rand() / RAND_MAX);
}
for (j = 0; j < size_B; j++)
{
h_B[j].re = ((float)rand() / RAND_MAX);
h_B[j].im = ((float)rand() / RAND_MAX);
}
}
//
void post_Correctness(Complex* h_C, Complex* h_C_chk, Complex* h_A, Complex* h_B, int size_idx_a, int size_idx_b, int size_idx_c, int size_idx_d, int size_idx_e, int size_idx_f)
{
// t3 [a,16,b,16,c,16,d,16] += sum(e,16,f,16) * t2 [a,e,b,f] * v2 [d,f,c,e];
int size_C = size_idx_a * size_idx_b * size_idx_c * size_idx_d;
long long int tmp_ops = 0;
int ops = 0;
int idx_a, idx_b, idx_c, idx_d, idx_e, idx_f;
for (idx_a = 0; idx_a < size_idx_a; idx_a++)
for (idx_b = 0; idx_b < size_idx_b; idx_b++)
for (idx_c = 0; idx_c < size_idx_c; idx_c++)
for (idx_d = 0; idx_d < size_idx_d; idx_d++)
{
int tmp_r_idx = idx_a + (idx_b + (idx_c + (idx_d) * size_idx_c) * size_idx_b) * size_idx_a;
#ifdef DEBUG_SIMPLE_CORRECTNESS
if (tmp_r_idx > 1024)
break;
#endif
for (idx_e = 0; idx_e < size_idx_e; idx_e++, ops = 0)
{
for (idx_f = 0; idx_f < size_idx_f; idx_f++)
{
h_C_chk[tmp_r_idx].re += (h_A[idx_a + (idx_e + (idx_b + (idx_f) * size_idx_b) * size_idx_e) * size_idx_a].re *
h_B[idx_d + (idx_f + (idx_c + (idx_e) * size_idx_c) * size_idx_f) * size_idx_d].re)
-(h_A[idx_a + (idx_e + (idx_b + (idx_f) * size_idx_b) * size_idx_e) * size_idx_a].im *
h_B[idx_d + (idx_f + (idx_c + (idx_e) * size_idx_c) * size_idx_f) * size_idx_d].im);
h_C_chk[tmp_r_idx].im += (h_A[idx_a + (idx_e + (idx_b + (idx_f) * size_idx_b) * size_idx_e) * size_idx_a].re *
h_B[idx_d + (idx_f + (idx_c + (idx_e) * size_idx_c) * size_idx_f) * size_idx_d].im)
+(h_A[idx_a + (idx_e + (idx_b + (idx_f) * size_idx_b) * size_idx_e) * size_idx_a].im *
h_B[idx_d + (idx_f + (idx_c + (idx_e) * size_idx_c) * size_idx_f) * size_idx_d].re);
ops++;
}
tmp_ops = tmp_ops + ops;
}
}
printf ("======================================= Correctness Check ==========================================\n");
float epsilon = 0.01;
int diff_re = 0;
int diff_im = 0;
int same_re = 0;
int same_im = 0;
int i;
for (i = 0; i < size_C; i++)
{
float check_re = h_C_chk[i].re - h_C[i].re;
float check_im = h_C_chk[i].im - h_C[i].im;
if (check_re < 0) check_re *= -1;
if (check_re > epsilon)
{
diff_re++;
if (diff_re < 8)
printf ("Index: %5d, (Host) %8.4f, (Dev.) %8.4f >> (Diff.) %8.4f\n", i, h_C_chk[i].re, h_C[i].re, check_re);
}
else
{
same_re++;
}
if (check_im < 0) check_im *= -1;
if (check_im > epsilon)
{
diff_im++;
if (diff_im < 8)
printf ("Index: %5d, (Host) %8.4f, (Dev.) %8.4f >> (Diff.) %8.4f\n", i, h_C_chk[i].im, h_C[i].im, check_im);
}
else
{
same_im++;
}
}
printf (" >>> PASSED on Re: %'10d among %'10d in t3\n", same_re, size_C);
printf (" >>> PASSED on Im: %'10d among %'10d in t3\n", same_im, size_C);
printf (" >>> ERROR on Re : %'10d among %'10d in t3\n", diff_re, size_C);
printf (" >>> ERROR on Im : %'10d among %'10d in t3\n", diff_im, size_C);
printf (" >>> Total Operations: %'lld\n", tmp_ops * 2);
printf ("====================================================================================================\n");
}
//
// t3 [a,16,b,16,c,16,d,16] += sum(e,16,f,16) * t2 [a,e,b,f] * v2 [d,f,c,e];
//
int main(int argc, char** argv)
{
// for sd2
Complex *host_C, *host_C_chk;
Complex *host_A;
Complex *host_B;
int size_idx_a, size_idx_b, size_idx_c, size_idx_d, size_idx_e, size_idx_f;
// Problem Size
size_idx_a = 16;
size_idx_b = 16;
size_idx_c = 16;
size_idx_d = 16;
size_idx_e = 16;
size_idx_f = 16;
//
if (argc == 7)
{
size_idx_a = atoi(argv[1]);
size_idx_b = atoi(argv[2]);
size_idx_c = atoi(argv[3]);
size_idx_d = atoi(argv[4]);
size_idx_e = atoi(argv[5]);
size_idx_f = atoi(argv[6]);
}
int size_C;
int size_A;
int size_B;
int size_internal;
// t3 [a,16,b,16,c,16,d,16] += sum(e,16,f,16) * t2 [a,e,b,f] * v2 [d,f,c,e];
size_internal = size_idx_e * size_idx_f;
size_C = size_idx_a * size_idx_b * size_idx_c * size_idx_d;
size_A = size_idx_a * size_idx_e * size_idx_b * size_idx_f;
size_B = size_idx_d * size_idx_f * size_idx_c * size_idx_e;
//
host_C = (Complex*)malloc(sizeof(Complex) * size_C);
host_C_chk = (Complex*)malloc(sizeof(Complex) * size_C);
host_A = (Complex*)malloc(sizeof(Complex) * size_A);
host_B = (Complex*)malloc(sizeof(Complex) * size_B);
printf ("==========================================================================================================\n");
printf (">>> abcd-aebf-dfce\n");
printf (">>> t3 [a,16,b,16,c,16,d,16] += sum(e,16,f,16) * t2 [a,e,b,f] * v2 [d,f,c,e];\n");
printf (">>> Problem Size (a,b,c,d) and (e,f): (%2d,%2d,%2d,%2d) and (%2d,%2d)\n", size_idx_a, size_idx_b, size_idx_c, size_idx_d, size_idx_e, size_idx_f);
printf ("==========================================================================================================\n");
// Initialze "1" Output and "2 x 9" Inputs
pre_Initializing_Input_Tensors(host_C, host_C_chk, size_C, host_A, size_A, host_B, size_B);
// Run the Kernels
sd_t_d2_fusion_(size_idx_a, size_idx_b, size_idx_c, size_idx_d, size_idx_e, size_idx_f, host_C, host_A, host_B, 1, -1);
#ifdef DEBUG_CORRECTNESS
// Correctness-Check
post_Correctness(host_C, host_C_chk, host_A, host_B, size_idx_a, size_idx_b, size_idx_c, size_idx_d, size_idx_e, size_idx_f);
#endif
// Free
free(host_C); free(host_C_chk);
free(host_A);
free(host_B);
return 0;
}
example/complex_contraction_3/CMakeLists.txt deleted 100644 → 0
View file @ 89f5d30a
add_example_executable(example_contraction_bilinear_xdl_fp32 contraction_bilinear_xdl_fp32.cpp)
add_example_executable(example_contraction_scale_xdl_fp32 contraction_scale_xdl_fp32.cpp)
add_example_executable(example_contraction_bilinear_xdl_fp64 contraction_bilinear_xdl_fp64.cpp)
add_example_executable(example_contraction_scale_xdl_fp64 contraction_scale_xdl_fp64.cpp)
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment