Files
rocm-systems/CMakeLists.txt
T
Jonathan R. Madsen 39f17ae8b8 rocm-smi and KokkosTools support (#23)
* renamed omnitrace_thread_data to thread_data

* initial implementation

* Numerous fixes and updates

- Updated timemory submodule
- Updated perfetto submodule (pulls in fixes for TRACE_EVENT)
- pthread_gotcha only after omnitrace_init_tooling
- omnitrace banner
- config settings for rocm-smi freq and devices
- critical_trace::get_entries
- OMNITRACE_BASIC_PRINT
- rocm_smi perfetto category
- redirect roctracer warnings for ROCm 4.5.0
- property specializations for rocm-smi components
- units fixes data_tracker types
- roctracer entries for pthread_create and start_thread
- omnitrace-avail defaults to settings, not components
- settings have conforming names
- settings warn about duplicates
- ptl named threads
- decreased max freq for sampler SIGALRM
- rocm-smi names thread
- rocm-smi avoids call to hipGetDeviceCount
- name roctracer activity callback threads
- fixed binary rewrite test output names

* Update lulesh example

- supports non-UVM GPU

* Lulesh tweaks + formatting

* KokkosP + Mode + Roctracer sampling deadlock fix

- kokkosp support
- omnitrace_init_library
- config::print_settings()
- config::get_mode()
- omnitrace::Mode
- omnitrace-avail improvements (removes settings)
- handle get_verbose() < 0
- disable dyninst InstrStackFrames by default
- handle perf_event_paranoid > 1 by disabling PAPI
- SIGALRM max freq to 5.0
- Name threads
- rocm-smi handles get_use_perfetto() and get_use_timemory()
- HSA_ENABLE_INTERRUPT=0 when roctracer + sampling (fixes deadlock)

* Tests, API renaming, roctracer

- disable renaming of thread 0
- verbprintf_bare
- enable dyninst merge tramp
- tweaked some omnitrace exe verbose levels
- reworked roctracer::setup and roctracer::shutdown
- rocm_smi::data::poll checks get_state()
- omnitrace_trace_finalize -> omnitrace_finalize
- omnitrace_trace_init -> omnitrace_init
- omnitrace_trace_set_env -> omnitrace_set_env
- omnitrace_trace_set_mpi -> omnitrace_set_mpi
- sampling mode does not disable timemory
- disable roctracer before shutting down rocm-smi
- lulesh tests w/ and w/o kokkosp
- lulesh tests for perfetto only
    - with --dynamic-callsites --traps --allow-overlapping
- lulesh tests for timemory only
    - with --stdlib --dynamic-callsites --traps --allow-overlapping

* Update timemory submodule

- fix for TIMEMORY_PROPERTY_SPECIALIZATION

* get_verbose() handling + timemory submodule update

- Findroctracer.cmake uses find_package(hsakmt)

* Stability fixes + rework roctracer + perfetto

- reworked roctracer start up
- critical_trace perfetto basic values
- perfetto sampling category
- sampler checks signals
- peak_rss in sampling
- pthread_gotcha::shutdown()
- rocm_smi::device_count()
- HSA_TOOLS_LIB is set
- HSA_ENABLE_INTERRUPT in omnitrace exe
- omnitrace exe verbosity level changes
- Avoid instrumenting Impl ns in Kokkos
- gpu::device_count prefers rocm_smi instead of hip
- ptl blocks signals
- fixed pthread_gotcha roctracer_data values
- removed runtime-instrument-sampling tests
- timemory submodule update

* cmake formatting

* timemory + roctracer updates

- fix timemory issue with papi_common
- fix timemory issue with units
- define roctracer::is_setup()

* Miscellaneous tweaks

- Disable sampling during runtime instrument
- Fixed warnings about dynamic callsites
- Fixed backtrace output when timemory disabled
- Test tweaks

* cmake-format

* omnitrace_target_compile_definitions

* timemory submodule update

* config, omnitrace, State, mpi_gotcha updates

- use OMNITRACE_THROW instead of direct throw
- is_attached()
- is_binary_rewrite()
- get_is_continuous_integration()
- get_debug_init()
- get_debug_finalize()
- max_thread_bookmarks default to 1
- State::Init
- app_thread oneTimeCode
- runtime instrumentation uses waitpid
- fixed init_names
- include main in MPI runs
- fixed sampling setup when disabled
- reworked mpi_gotcha
- disabled critical trace in transpose test

* cmake-format

* handle rocm_smi::device_count() exception

* CI timeouts

* Re-enable runtime-instrument + sampling
2022-02-08 17:42:17 -06:00

380 wiersze
14 KiB
CMake

cmake_minimum_required(VERSION 3.15 FATAL_ERROR)
if(CMAKE_SOURCE_DIR STREQUAL CMAKE_BINARY_DIR AND CMAKE_CURRENT_SOURCE_DIR STREQUAL
CMAKE_SOURCE_DIR)
set(MSG "")
message(STATUS "Warning! Building from the source directory is not recommended")
message(STATUS "If unintented, please remove 'CMakeCache.txt' and 'CMakeFiles'")
message(STATUS "and build from a separate directory")
message(AUTHOR_WARNING "In-source build")
endif()
if(NOT UNIX OR APPLE)
message(
AUTHOR_WARNING
"omnitrace only supports Linux. Configure and/or build is likely to fail")
endif()
file(READ "${CMAKE_CURRENT_SOURCE_DIR}/VERSION" FULL_VERSION_STRING LIMIT_COUNT 1)
string(REGEX REPLACE "(\n|\r)" "" FULL_VERSION_STRING "${FULL_VERSION_STRING}")
string(REGEX REPLACE "([0-9]+)\.([0-9]+)\.([0-9]+)(.*)" "\\1.\\2.\\3" OMNITRACE_VERSION
"${FULL_VERSION_STRING}")
project(
omnitrace
LANGUAGES C CXX
VERSION ${OMNITRACE_VERSION}
DESCRIPTION "CPU/GPU Application tracing with static/dynamic binary instrumentation"
HOMEPAGE_URL "https://github.com/AARInternal/omnitrace")
message(
STATUS
"[${PROJECT_NAME}] version ${PROJECT_VERSION_MAJOR}.${PROJECT_VERSION_MINOR}.${PROJECT_VERSION_PATCH}"
)
set(CMAKE_MODULE_PATH ${PROJECT_SOURCE_DIR}/cmake ${PROJECT_SOURCE_DIR}/cmake/Modules
${CMAKE_MODULE_PATH})
set(BUILD_SHARED_LIBS
ON
CACHE BOOL "Build shared libraries")
set(BUILD_STATIC_LIBS
OFF
CACHE BOOL "Build static libraries")
set(CMAKE_POSITION_INDEPENDENT_CODE
ON
CACHE BOOL "Build position independent code")
if("${CMAKE_BUILD_TYPE}" STREQUAL "")
set(CMAKE_BUILD_TYPE
RelWithDebInfo
CACHE STRING "Build type" FORCE)
endif()
include(GNUInstallDirs) # install directories
include(MacroUtilities) # various functions and macros
include(Compilers) # compiler identification
include(BuildSettings) # compiler flags
set(CMAKE_CXX_STANDARD
17
CACHE STRING "CXX language standard")
omnitrace_add_feature(CMAKE_CXX_STANDARD "CXX language standard")
omnitrace_add_feature(CMAKE_BUILD_TYPE "Build optimization level")
omnitrace_add_option(CMAKE_CXX_STANDARD_REQUIRED "Require C++ language standard" ON)
omnitrace_add_option(CMAKE_CXX_EXTENSIONS "Compiler specific language extensions" OFF)
omnitrace_add_option(CMAKE_INSTALL_RPATH_USE_LINK_PATH "Enable rpath to linked libraries"
ON)
omnitrace_add_option(OMNITRACE_USE_CLANG_TIDY "Enable clang-tidy" OFF)
omnitrace_add_option(OMNITRACE_USE_MPI "Enable MPI support" OFF)
omnitrace_add_option(OMNITRACE_USE_HIP "Enable HIP support" ON)
omnitrace_add_option(OMNITRACE_USE_ROCTRACER "Enable roctracer support"
${OMNITRACE_USE_HIP})
omnitrace_add_option(
OMNITRACE_USE_ROCM_SMI "Enable rocm-smi support for power/temp/etc. sampling"
${OMNITRACE_USE_HIP})
omnitrace_add_option(OMNITRACE_USE_MPI_HEADERS
"Enable wrapping MPI functions w/o enabling MPI dependency" OFF)
omnitrace_add_option(OMNITRACE_BUILD_DYNINST "Build dyninst from submodule" OFF)
omnitrace_add_option(OMNITRACE_BUILD_TESTING "Enable building the testing suite" OFF)
omnitrace_add_option(OMNITRACE_CUSTOM_DATA_SOURCE "Enable custom data source" OFF)
omnitrace_add_option(OMNITRACE_BUILD_HIDDEN_VISIBILITY
"Build with hidden visibility (disable for Debug builds)" ON)
if(NOT OMNITRACE_USE_HIP)
set(OMNITRACE_USE_ROCTRACER
OFF
CACHE BOOL "Disabled via OMNITRACE_USE_HIP=OFF" FORCE)
endif()
include(ProcessorCount)
processorcount(OMNITRACE_PROCESSOR_COUNT)
math(EXPR OMNITRACE_THREAD_COUNT "16 * ${OMNITRACE_PROCESSOR_COUNT}")
if(OMNITRACE_THREAD_COUNT LESS 128)
set(OMNITRACE_THREAD_COUNT 128)
endif()
set(OMNITRACE_MAX_THREADS
"${OMNITRACE_THREAD_COUNT}"
CACHE
STRING
"Maximum number of threads in the host application. Likely only needs to be increased if host app does not use thread-pool but creates many threads"
)
omnitrace_add_feature(
OMNITRACE_MAX_THREADS
"Maximum number of total threads supported in the host application (default: max of 128 or 16 * nproc)"
)
# default visibility settings
set(CMAKE_C_VISIBILITY_PRESET
"default"
CACHE STRING "Visibility preset for non-inline C functions")
set(CMAKE_CXX_VISIBILITY_PRESET
"default"
CACHE STRING "Visibility preset for non-inline C++ functions/objects")
set(CMAKE_VISIBILITY_INLINES_HIDDEN
OFF
CACHE BOOL "Visibility preset for inline functions")
set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
include(Formatting) # format target
include(Packages) # finds third-party libraries
if(OMNITRACE_USE_HIP OR OMNITRACE_USE_ROCTRACER)
find_package(HIP QUIET)
if(HIP_VERSION_MAJOR GREATER_EQUAL 4 AND HIP_VERSION_MINOR GREATER 3)
set(roctracer_kfdwrapper_LIBRARY)
endif()
else()
set(HIP_VERSION "0.0.0")
set(HIP_VERSION_MAJOR 0)
set(HIP_VERSION_MINOR 0)
set(HIP_VERSION_PATCH 0)
endif()
configure_file(${PROJECT_SOURCE_DIR}/include/library/defines.hpp.in
${PROJECT_BINARY_DIR}/include/library/defines.hpp @ONLY)
omnitrace_activate_clang_tidy()
# custom visibility settings
if(OMNITRACE_BUILD_HIDDEN_VISIBILITY)
set(CMAKE_C_VISIBILITY_PRESET "hidden")
set(CMAKE_CXX_VISIBILITY_PRESET "hidden")
set(CMAKE_VISIBILITY_INLINES_HIDDEN ON)
endif()
if(NOT DEFINED CMAKE_INTERPROCEDURAL_OPTIMIZATION)
set(CMAKE_INTERPROCEDURAL_OPTIMIZATION OFF)
endif()
if(OMNITRACE_BUILD_LTO)
omnitrace_save_variables(LTO VARIABLES CMAKE_INTERPROCEDURAL_OPTIMIZATION)
set(CMAKE_INTERPROCEDURAL_OPTIMIZATION ON)
endif()
# ------------------------------------------------------------------------------#
#
# omnitrace-library target
#
# ------------------------------------------------------------------------------#
set(library_sources
${CMAKE_CURRENT_LIST_DIR}/src/library.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/config.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/critical_trace.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/kokkosp.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/gpu.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/perfetto.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/ptl.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/sampling.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/state.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/thread_data.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/timemory.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/components/backtrace.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/components/fork_gotcha.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/components/mpi_gotcha.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/components/omnitrace.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/components/pthread_gotcha.cpp
${perfetto_DIR}/sdk/perfetto.cc)
set(library_headers
${CMAKE_CURRENT_LIST_DIR}/include/library.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/api.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/config.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/common.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/critical_trace.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/debug.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/gpu.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/perfetto.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/ptl.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/sampling.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/state.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/thread_data.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/timemory.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/fwd.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/backtrace.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/fork_gotcha.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/mpi_gotcha.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/omnitrace.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/rocm_smi.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/roctracer.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/roctracer_callbacks.hpp
${CMAKE_CURRENT_LIST_DIR}/include/library/components/pthread_gotcha.hpp
${perfetto_DIR}/sdk/perfetto.h)
add_library(omnitrace-library SHARED ${library_sources} ${library_headers})
if(OMNITRACE_USE_ROCTRACER)
target_sources(
omnitrace-library
PRIVATE ${CMAKE_CURRENT_LIST_DIR}/src/library/components/roctracer.cpp
${CMAKE_CURRENT_LIST_DIR}/src/library/components/roctracer_callbacks.cpp)
endif()
if(OMNITRACE_USE_ROCM_SMI)
target_sources(omnitrace-library
PRIVATE ${CMAKE_CURRENT_LIST_DIR}/src/library/components/rocm_smi.cpp)
endif()
target_include_directories(omnitrace-library SYSTEM PRIVATE ${perfetto_DIR}/sdk)
target_compile_definitions(
omnitrace-library
PRIVATE OMNITRACE_MAX_THREADS=${OMNITRACE_MAX_THREADS}
$<IF:$<BOOL:${OMNITRACE_CUSTOM_DATA_SOURCE}>,CUSTOM_DATA_SOURCE,>)
target_link_libraries(
omnitrace-library
PUBLIC $<BUILD_INTERFACE:omnitrace::omnitrace-headers>
$<BUILD_INTERFACE:omnitrace::omnitrace-threading>
$<BUILD_INTERFACE:omnitrace::omnitrace-compile-options>
$<BUILD_INTERFACE:omnitrace::omnitrace-hip>
$<BUILD_INTERFACE:omnitrace::omnitrace-roctracer>
$<BUILD_INTERFACE:omnitrace::omnitrace-rocm-smi>
$<BUILD_INTERFACE:omnitrace::omnitrace-mpi>
$<BUILD_INTERFACE:omnitrace::omnitrace-ptl>
$<BUILD_INTERFACE:timemory::timemory-headers>
$<BUILD_INTERFACE:timemory::timemory-gotcha>
$<BUILD_INTERFACE:timemory::timemory-cxx-shared>
$<IF:$<BOOL:${OMNITRACE_USE_SANITIZER}>,omnitrace::omnitrace-sanitizer,>)
if(OMNITRACE_DYNINST_API_RT)
get_filename_component(OMNITRACE_DYNINST_API_RT_DIR "${OMNITRACE_DYNINST_API_RT}"
DIRECTORY)
endif()
set_target_properties(
omnitrace-library
PROPERTIES OUTPUT_NAME omnitrace
INSTALL_RPATH
"\$ORIGIN:\$ORIGIN/timemory/libunwind:\$ORIGIN/dyninst-tpls/libs")
install(
TARGETS omnitrace-library
DESTINATION ${CMAKE_INSTALL_LIBDIR}
OPTIONAL)
# ------------------------------------------------------------------------------#
#
# omnitrace-avail target
#
# ------------------------------------------------------------------------------#
add_executable(omnitrace-avail ${CMAKE_CURRENT_LIST_DIR}/src/avail.cpp
${CMAKE_CURRENT_LIST_DIR}/include/avail.hpp)
target_include_directories(omnitrace-avail PRIVATE ${CMAKE_CURRENT_LIST_DIR}/include)
target_compile_definitions(omnitrace-avail PRIVATE OMNITRACE_EXTERN_COMPONENTS=0)
target_link_libraries(omnitrace-avail PRIVATE omnitrace-library)
set_target_properties(omnitrace-avail PROPERTIES INSTALL_RPATH_USE_LINK_PATH ON)
install(
TARGETS omnitrace-avail
DESTINATION bin
OPTIONAL)
# ------------------------------------------------------------------------------#
#
# omnitrace-exe target
#
# ------------------------------------------------------------------------------#
add_executable(
omnitrace-exe
${_EXCLUDE} ${CMAKE_CURRENT_LIST_DIR}/src/omnitrace.cpp
${CMAKE_CURRENT_LIST_DIR}/include/omnitrace.hpp
${CMAKE_CURRENT_LIST_DIR}/src/omnitrace/details.cpp)
target_link_libraries(
omnitrace-exe
PRIVATE omnitrace::omnitrace-headers
omnitrace::omnitrace-dyninst
omnitrace::omnitrace-compile-options
$<BUILD_INTERFACE:timemory::timemory-headers>
$<IF:$<BOOL:${OMNITRACE_USE_SANITIZER}>,omnitrace::omnitrace-sanitizer,>)
set_target_properties(
omnitrace-exe
PROPERTIES
OUTPUT_NAME omnitrace
INSTALL_RPATH_USE_LINK_PATH ON
INSTALL_RPATH
"\$ORIGIN/../${CMAKE_INSTALL_LIBDIR}:\$ORIGIN/../${CMAKE_INSTALL_LIBDIR}/timemory/libunwind:\$ORIGIN/../${CMAKE_INSTALL_LIBDIR}/dyninst-tpls/lib"
)
if(CMAKE_BUILD_TYPE MATCHES "^(DEBUG|Debug)")
string(REPLACE " " ";" _FLAGS "${CMAKE_CXX_FLAGS_RELWITHDEBINFO}")
target_compile_options(omnitrace-exe PRIVATE ${_FLAGS})
endif()
install(
TARGETS omnitrace-exe
DESTINATION ${CMAKE_INSTALL_BINDIR}
OPTIONAL)
# ------------------------------------------------------------------------------#
#
# miscellaneous installs
#
# ------------------------------------------------------------------------------#
configure_file(${PROJECT_SOURCE_DIR}/scripts/setup-env.sh.in
${PROJECT_BINARY_DIR}/scripts/setup-env.sh @ONLY)
install(
PROGRAMS ${PROJECT_SOURCE_DIR}/scripts/omnitrace-merge.jl
DESTINATION ${CMAKE_INSTALL_BINDIR}
OPTIONAL)
install(
FILES ${PROJECT_SOURCE_DIR}/roctrace.cfg
DESTINATION ${CMAKE_INSTALL_DATAROOTDIR}/${PROJECT_NAME}
OPTIONAL)
install(
FILES ${PROJECT_BINARY_DIR}/scripts/setup-env.sh
DESTINATION ${CMAKE_INSTALL_DATAROOTDIR}/${PROJECT_NAME}
OPTIONAL)
# ------------------------------------------------------------------------------#
#
# examples
#
# ------------------------------------------------------------------------------#
if(OMNITRACE_BUILD_LTO)
omnitrace_restore_variables(LTO VARIABLES CMAKE_INTERPROCEDURAL_OPTIMIZATION)
endif()
if(OMNITRACE_BUILD_TESTING)
add_subdirectory(examples)
endif()
# ------------------------------------------------------------------------------#
#
# tests
#
# ------------------------------------------------------------------------------#
if(OMNITRACE_BUILD_TESTING)
include(CTest)
enable_testing()
add_subdirectory(tests)
endif()
# ------------------------------------------------------------------------------#
#
# packaging
#
# ------------------------------------------------------------------------------#
include(ConfigCPack)
# ------------------------------------------------------------------------------#
#
# config info
#
# ------------------------------------------------------------------------------#
omnitrace_print_features()