d496bcef18
* Fix dimension mismatch for multi-GPU systems with identical architectures This change addresses an issue where counter dimensions were incorrectly shared across all GPU agents with the same architecture name, even when those agents had different hardware configurations (e.g., different CU counts). Changes: - Updated getBlockDimensions() to accept agent ID instead of architecture name - Made dimension cache agent-specific instead of architecture-specific - Updated set_dimensions() in AST evaluation to use specific agent ID - Modified all API functions to handle agent-specific dimension lookups - Updated tests to work with agent-specific dimensions This fix ensures that dimensions accurately reflect the actual hardware configuration of each individual GPU agent, preventing dimension mismatches in multi-GPU systems where GPUs share the same architecture but have different physical configurations. Counter ID Representation Changes: - Modified counter_id encoding to include agent information in bits 37-32 - Agent logical_node_id is encoded as (value + 1) to ensure agent 0 is detectable - Counter records internally store only 16-bit base metric IDs (bits 15-0) - Tool reconstructs agent-encoded counter IDs from base metric ID & agent info - Instance record counter_id field uses bitwise AND mask to extract base metric ID (counter_id.handle & 0xFFFF) to fit in 16-bit storage - Output generators (CSV, JSON, Perfetto) use agent-encoded IDs for consistency - Updated counter_config.cpp and metrics.cpp to extract base metric ID when needed - All counter lookups now properly handle agent-encoded vs base metric IDs This ensures counter IDs are consistent between metadata and output records while maintaining compact storage in instance records.
155 regels
5.0 KiB
CMake
155 regels
5.0 KiB
CMake
#
|
|
#
|
|
#
|
|
cmake_minimum_required(VERSION 3.21.0 FATAL_ERROR)
|
|
|
|
if(NOT CMAKE_HIP_COMPILER)
|
|
find_program(
|
|
amdclangpp_EXECUTABLE
|
|
NAMES amdclang++
|
|
HINTS ${ROCM_PATH} ENV ROCM_PATH /opt/rocm
|
|
PATHS ${ROCM_PATH} ENV ROCM_PATH /opt/rocm
|
|
PATH_SUFFIXES bin llvm/bin NO_CACHE)
|
|
mark_as_advanced(amdclangpp_EXECUTABLE)
|
|
|
|
if(amdclangpp_EXECUTABLE)
|
|
set(CMAKE_HIP_COMPILER "${amdclangpp_EXECUTABLE}")
|
|
endif()
|
|
endif()
|
|
|
|
project(
|
|
rocprofiler-sdk-tests-thread-trace
|
|
LANGUAGES CXX HIP
|
|
VERSION 0.0.0)
|
|
|
|
find_package(rocprofiler-sdk REQUIRED)
|
|
|
|
set(CMAKE_CXX_STANDARD 17)
|
|
set(CMAKE_CXX_EXTENSIONS OFF)
|
|
set(CMAKE_CXX_STANDARD_REQUIRED ON)
|
|
set(CMAKE_HIP_STANDARD 17)
|
|
set(CMAKE_HIP_EXTENSIONS OFF)
|
|
set(CMAKE_HIP_STANDARD_REQUIRED ON)
|
|
|
|
foreach(_TYPE DEBUG MINSIZEREL RELEASE RELWITHDEBINFO)
|
|
if("${CMAKE_HIP_FLAGS_${_TYPE}}" STREQUAL "")
|
|
set(CMAKE_HIP_FLAGS_${_TYPE} "${CMAKE_CXX_FLAGS_${_TYPE}}")
|
|
endif()
|
|
endforeach()
|
|
|
|
set_source_files_properties(
|
|
kernel_branch.cpp kernel_lds.cpp
|
|
PROPERTIES LANGUAGE HIP COMPILE_FLAGS "${CMAKE_HIP_FLAGS_RELWITHDEBINFO}")
|
|
set_source_files_properties(main.cpp PROPERTIES LANGUAGE HIP)
|
|
|
|
# Single dispatch test
|
|
add_executable(thread-trace-api-single-test)
|
|
target_sources(
|
|
thread-trace-api-single-test PRIVATE main.cpp trace_callbacks.cpp single_dispatch.cpp
|
|
kernel_branch.cpp kernel_lds.cpp)
|
|
target_link_libraries(
|
|
thread-trace-api-single-test
|
|
PRIVATE rocprofiler-sdk::rocprofiler-sdk rocprofiler-sdk::tests-build-flags
|
|
rocprofiler-sdk::tests-common-library)
|
|
|
|
add_test(NAME thread-trace-api-single-test
|
|
COMMAND $<TARGET_FILE:thread-trace-api-single-test>)
|
|
|
|
set_tests_properties(
|
|
thread-trace-api-single-test
|
|
PROPERTIES TIMEOUT
|
|
10
|
|
LABELS
|
|
"integration-tests"
|
|
ENVIRONMENT
|
|
"${ROCPROFILER_MEMCHECK_PRELOAD_ENV}"
|
|
FAIL_REGULAR_EXPRESSION
|
|
"${ROCPROFILER_DEFAULT_FAIL_REGEX}"
|
|
DISABLED
|
|
${ROCPROFILER_DISABLE_UNSTABLE_CTESTS})
|
|
|
|
# Multi dispatch test
|
|
add_executable(thread-trace-api-multi-test)
|
|
target_sources(
|
|
thread-trace-api-multi-test PRIVATE main.cpp trace_callbacks.cpp multi_dispatch.cpp
|
|
kernel_branch.cpp kernel_lds.cpp)
|
|
target_link_libraries(
|
|
thread-trace-api-multi-test
|
|
PRIVATE rocprofiler-sdk::rocprofiler-sdk rocprofiler-sdk::tests-build-flags
|
|
rocprofiler-sdk::tests-common-library)
|
|
|
|
add_test(NAME thread-trace-api-multi-test
|
|
COMMAND $<TARGET_FILE:thread-trace-api-multi-test>)
|
|
|
|
set_tests_properties(
|
|
thread-trace-api-multi-test
|
|
PROPERTIES TIMEOUT
|
|
10
|
|
LABELS
|
|
"integration-tests"
|
|
ENVIRONMENT
|
|
"${ROCPROFILER_MEMCHECK_PRELOAD_ENV}"
|
|
FAIL_REGULAR_EXPRESSION
|
|
"${ROCPROFILER_DEFAULT_FAIL_REGEX}"
|
|
DISABLED
|
|
${ROCPROFILER_DISABLE_UNSTABLE_CTESTS})
|
|
|
|
# Agent profiling test
|
|
add_executable(thread-trace-api-agent-test)
|
|
target_sources(thread-trace-api-agent-test PRIVATE main.cpp trace_callbacks.cpp agent.cpp
|
|
kernel_branch.cpp kernel_lds.cpp)
|
|
target_link_libraries(
|
|
thread-trace-api-agent-test
|
|
PRIVATE rocprofiler-sdk::rocprofiler-sdk rocprofiler-sdk::tests-build-flags
|
|
rocprofiler-sdk::tests-common-library)
|
|
|
|
add_test(NAME thread-trace-api-agent-test
|
|
COMMAND $<TARGET_FILE:thread-trace-api-agent-test>)
|
|
|
|
set_tests_properties(
|
|
thread-trace-api-agent-test
|
|
PROPERTIES TIMEOUT
|
|
10
|
|
LABELS
|
|
"integration-tests"
|
|
ENVIRONMENT
|
|
"${ROCPROFILER_MEMCHECK_PRELOAD_ENV}"
|
|
FAIL_REGULAR_EXPRESSION
|
|
"${ROCPROFILER_DEFAULT_FAIL_REGEX}"
|
|
DISABLED
|
|
${ROCPROFILER_DISABLE_UNSTABLE_CTESTS})
|
|
|
|
# Test large buffer sizes. 5120 == 5GB
|
|
add_test(NAME thread-trace-api-large-buffer-test
|
|
COMMAND $<TARGET_FILE:thread-trace-api-agent-test>)
|
|
|
|
set_tests_properties(
|
|
thread-trace-api-large-buffer-test
|
|
PROPERTIES TIMEOUT
|
|
10
|
|
LABELS
|
|
"integration-tests"
|
|
ENVIRONMENT
|
|
"${PRELOAD_ENV};ATT_BUFFER_SIZE_MB=5120"
|
|
FAIL_REGULAR_EXPRESSION
|
|
"${ROCPROFILER_DEFAULT_FAIL_REGEX}"
|
|
DISABLED
|
|
${ROCPROFILER_DISABLE_UNSTABLE_CTESTS})
|
|
|
|
# Test occupancy mode
|
|
add_test(NAME thread-trace-api-extra-args
|
|
COMMAND $<TARGET_FILE:thread-trace-api-agent-test>)
|
|
|
|
set_tests_properties(
|
|
thread-trace-api-extra-args
|
|
PROPERTIES TIMEOUT
|
|
10
|
|
LABELS
|
|
"integration-tests"
|
|
ENVIRONMENT
|
|
"${PRELOAD_ENV};ATT_NODETAIL=1"
|
|
FAIL_REGULAR_EXPRESSION
|
|
"${ROCPROFILER_DEFAULT_FAIL_REGEX}"
|
|
DISABLED
|
|
${ROCPROFILER_DISABLE_UNSTABLE_CTESTS})
|