AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Remove Critical Trace Support (#327)

Browse Source 
* Delete core critical-trace files

* Update docs and README

* Update workflows

* Update testing

* Update cmake

* Remove critical trace usage in source code

* Update source/docs/critical_trace.md

- fix spelling

* Formatting

* Update bin/omnitrace-avail/avail.cpp

- statically allocate shared pointers for timemory manager and hash id/aliases to prevent use-after-free errors
This commit is contained in:
Jonathan R. Madsen
2024-04-23 09:35:44 -05:00
committed by GitHub
parent b81db80926
commit 9499e2f521
45 changed files with 55 additions and 2954 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/CMakeLists.txt
-1
View File
@@ -17,7 +17,6 @@ include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-openmp-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-code-coverage-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-fork-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-time-window-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-critical-trace-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-attach-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-rccl-tests.cmake)
include(${CMAKE_CURRENT_LIST_DIR}/omnitrace-overflow-tests.cmake)
tests/omnitrace-critical-trace-tests.cmake
-54
View File
@@ -1,54 +0,0 @@
# -------------------------------------------------------------------------------------- #
#
# critical-trace tests
#
# -------------------------------------------------------------------------------------- #
omnitrace_add_test(
SKIP_BASELINE SKIP_RUNTIME SKIP_SAMPLING
NAME parallel-overhead-critical-trace
TARGET parallel-overhead
LABELS "critical-trace"
REWRITE_ARGS
-e
-i
8
-E
"^fib"
-v
2
--print-instrumented
functions
RUN_ARGS 10 4 100
ENVIRONMENT "${_critical_trace_environment}")
add_test(
NAME parallel-overhead-process-critical-trace
COMMAND
$<TARGET_FILE:omnitrace-critical-trace>
${PROJECT_BINARY_DIR}/omnitrace-tests-output/parallel-overhead-critical-trace-binary-rewrite/call-chain.json
WORKING_DIRECTORY ${PROJECT_BINARY_DIR})
set(_parallel_overhead_critical_trace_environ
"OMNITRACE_OUTPUT_PATH=omnitrace-tests-output"
"OMNITRACE_OUTPUT_PREFIX=parallel-overhead-critical-trace/"
"OMNITRACE_CRITICAL_TRACE_DEBUG=ON"
"OMNITRACE_VERBOSE=4"
"OMNITRACE_USE_PID=OFF"
"OMNITRACE_TIME_OUTPUT=OFF"
"OMNITRACE_CI=ON"
"OMNITRACE_CI_TIMEOUT=300")
set_tests_properties(
parallel-overhead-process-critical-trace
PROPERTIES
ENVIRONMENT
"${_parallel_overhead_critical_trace_environ}"
TIMEOUT
300
LABELS
"parallel-overhead;critical-trace"
PASS_REGULAR_EXPRESSION
"Outputting.*(critical-trace-cpu.json).*Outputting.*(critical-trace-any.json)"
DEPENDS
parallel-overhead-critical-trace-binary-rewrite-run)
tests/omnitrace-fork-tests.cmake
+1 -1
View File
@@ -10,7 +10,7 @@ omnitrace_add_test(
REWRITE_ARGS -e -v 2 --print-instrumented modules -i 16
RUNTIME_ARGS -e -v 1 --label file -i 16
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=ON;OMNITRACE_SAMPLING_FREQ=250;OMNITRACE_SAMPLING_REALTIME=ON"
"${_base_environment};OMNITRACE_SAMPLING_FREQ=250;OMNITRACE_SAMPLING_REALTIME=ON"
SAMPLING_PASS_REGEX "fork.. called on PID"
RUNTIME_PASS_REGEX "fork.. called on PID"
REWRITE_RUN_PASS_REGEX "fork.. called on PID"
tests/omnitrace-instrument-tests.cmake
+1 -1
View File
@@ -40,7 +40,7 @@ omnitrace_add_test(
return
args
RUN_ARGS 10 ${NUM_THREADS} 1000
ENVIRONMENT "${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF")
ENVIRONMENT "${_base_environment}")
omnitrace_add_test(
SKIP_BASELINE SKIP_RUNTIME
tests/omnitrace-kokkos-tests.cmake
+6 -9
View File
@@ -26,7 +26,7 @@ omnitrace_add_test(
LABELS "kokkos;kokkos-profile-library"
RUN_ARGS -i 25 -s 20 -p
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_KOKKOSP=ON;OMNITRACE_COUT_OUTPUT=ON;OMNITRACE_SAMPLING_FREQ=50;OMNITRACE_KOKKOSP_PREFIX=[kokkos];KOKKOS_PROFILE_LIBRARY=libomnitrace-dl.so"
"${_base_environment};OMNITRACE_USE_KOKKOSP=ON;OMNITRACE_COUT_OUTPUT=ON;OMNITRACE_SAMPLING_FREQ=50;OMNITRACE_KOKKOSP_PREFIX=[kokkos];KOKKOS_PROFILE_LIBRARY=libomnitrace-dl.so"
REWRITE_RUN_PASS_REGEX "\\|_\\[kokkos\\] [a-zA-Z]"
RUNTIME_PASS_REGEX "\\|_\\[kokkos\\] [a-zA-Z]")
@@ -40,7 +40,7 @@ omnitrace_add_test(
LABELS "kokkos;kokkos-profile-library"
RUN_ARGS -i 10 -s 20 -p
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_KOKKOSP=ON;OMNITRACE_COUT_OUTPUT=ON;OMNITRACE_SAMPLING_FREQ=50;OMNITRACE_KOKKOSP_PREFIX=[kokkos];KOKKOS_PROFILE_LIBRARY=libomnitrace.so"
"${_base_environment};OMNITRACE_USE_KOKKOSP=ON;OMNITRACE_COUT_OUTPUT=ON;OMNITRACE_SAMPLING_FREQ=50;OMNITRACE_KOKKOSP_PREFIX=[kokkos];KOKKOS_PROFILE_LIBRARY=libomnitrace.so"
BASELINE_PASS_REGEX "\\|_\\[kokkos\\] [a-zA-Z]")
omnitrace_add_test(
@@ -53,7 +53,7 @@ omnitrace_add_test(
LABELS "kokkos;kokkos-profile-library"
RUN_ARGS -i 10 -s 20 -p
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_KOKKOSP=ON;OMNITRACE_COUT_OUTPUT=ON;OMNITRACE_SAMPLING_FREQ=50;OMNITRACE_KOKKOSP_PREFIX=[kokkos];KOKKOS_PROFILE_LIBRARY=libomnitrace-dl.so"
"${_base_environment};OMNITRACE_USE_KOKKOSP=ON;OMNITRACE_COUT_OUTPUT=ON;OMNITRACE_SAMPLING_FREQ=50;OMNITRACE_KOKKOSP_PREFIX=[kokkos];KOKKOS_PROFILE_LIBRARY=libomnitrace-dl.so"
BASELINE_PASS_REGEX "\\|_\\[kokkos\\] [a-zA-Z]")
omnitrace_add_test(
@@ -77,8 +77,7 @@ omnitrace_add_test(
-ME
[==[lib(gomp|m-)]==]
RUN_ARGS -i 10 -s 20 -p
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_KOKKOSP=ON")
ENVIRONMENT "${_base_environment};OMNITRACE_USE_KOKKOSP=ON")
omnitrace_add_test(
SKIP_BASELINE
@@ -100,8 +99,7 @@ omnitrace_add_test(
-ME
[==[libgomp]==]
RUN_ARGS -i 10 -s 20 -p
ENVIRONMENT
"${_perfetto_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_KOKKOSP=OFF")
ENVIRONMENT "${_perfetto_environment};OMNITRACE_USE_KOKKOSP=OFF")
omnitrace_add_test(
NAME lulesh-timemory
@@ -122,6 +120,5 @@ omnitrace_add_test(
--env
OMNITRACE_TIMEMORY_COMPONENTS="wall_clock peak_rss"
RUN_ARGS -i 10 -s 20 -p
ENVIRONMENT
"${_timemory_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_KOKKOSP=OFF"
ENVIRONMENT "${_timemory_environment};OMNITRACE_USE_KOKKOSP=OFF"
REWRITE_FAIL_REGEX "0 instrumented loops in procedure")
tests/omnitrace-rocm-tests.cmake
+5 -5
View File
@@ -26,7 +26,7 @@ omnitrace_add_test(
args
-E
uniform_int_distribution
ENVIRONMENT "${_base_environment};OMNITRACE_CRITICAL_TRACE=ON")
ENVIRONMENT "${_base_environment}")
omnitrace_add_test(
SKIP_REWRITE SKIP_RUNTIME
@@ -37,7 +37,7 @@ omnitrace_add_test(
NUM_PROCS 1
RUN_ARGS 1 2 2
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_ROCTRACER_HSA_ACTIVITY=OFF;OMNITRACE_ROCTRACER_HSA_API=OFF"
"${_base_environment};OMNITRACE_ROCTRACER_HSA_ACTIVITY=OFF;OMNITRACE_ROCTRACER_HSA_API=OFF"
)
omnitrace_add_test(
@@ -61,7 +61,7 @@ omnitrace_add_test(
-E
uniform_int_distribution
RUN_ARGS 2 100 50
ENVIRONMENT "${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF"
ENVIRONMENT "${_base_environment}"
REWRITE_FAIL_REGEX "0 instrumented loops in procedure transpose")
if(OMNITRACE_USE_ROCPROFILER)
@@ -75,7 +75,7 @@ if(OMNITRACE_USE_ROCPROFILER)
NUM_PROCS ${NUM_PROCS}
REWRITE_ARGS -e -v 2 -E uniform_int_distribution
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_ROCM_EVENTS=${OMNITRACE_ROCM_EVENTS_TEST}"
"${_base_environment};OMNITRACE_ROCM_EVENTS=${OMNITRACE_ROCM_EVENTS_TEST}"
REWRITE_RUN_PASS_REGEX
"rocprof-device-0-GRBM_COUNT.txt(.*)rocprof-device-0-GPUBusy.txt(.*)rocprof-device-0-SQ_WAVES.txt(.*)rocprof-device-0-SQ_INSTS_VALU.txt(.*)rocprof-device-0-VALUInsts.txt(.*)rocprof-device-0-TCC_HIT_sum.txt(.*)rocprof-device-0-TA_TA_BUSY_0.txt(.*)rocprof-device-0-TA_TA_BUSY_11.txt"
)
@@ -90,7 +90,7 @@ if(OMNITRACE_USE_ROCPROFILER)
NUM_PROCS ${NUM_PROCS}
REWRITE_ARGS -e -v 2 -E uniform_int_distribution
ENVIRONMENT
"${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF;OMNITRACE_USE_ROCTRACER=OFF;OMNITRACE_ROCM_EVENTS=${OMNITRACE_ROCM_EVENTS_TEST}"
"${_base_environment};OMNITRACE_USE_ROCTRACER=OFF;OMNITRACE_ROCM_EVENTS=${OMNITRACE_ROCM_EVENTS_TEST}"
REWRITE_RUN_PASS_REGEX
"rocprof-device-0-GRBM_COUNT.txt(.*)rocprof-device-0-GPUBusy.txt(.*)rocprof-device-0-SQ_WAVES.txt(.*)rocprof-device-0-SQ_INSTS_VALU.txt(.*)rocprof-device-0-VALUInsts.txt(.*)rocprof-device-0-TCC_HIT_sum.txt(.*)rocprof-device-0-TA_TA_BUSY_0.txt(.*)rocprof-device-0-TA_TA_BUSY_11.txt"
REWRITE_RUN_FAIL_REGEX "roctracer.txt|OMNITRACE_ABORT_FAIL_REGEX")
tests/omnitrace-testing.cmake
+1 -18
View File
@@ -80,7 +80,6 @@ set(_lock_environment
"OMNITRACE_USE_SAMPLING=ON"
"OMNITRACE_USE_PROCESS_SAMPLING=OFF"
"OMNITRACE_SAMPLING_FREQ=750"
"OMNITRACE_CRITICAL_TRACE=ON"
"OMNITRACE_COLLAPSE_THREADS=ON"
"OMNITRACE_TRACE_THREAD_LOCKS=ON"
"OMNITRACE_TRACE_THREAD_SPIN_LOCKS=ON"
@@ -91,26 +90,11 @@ set(_lock_environment
"OMNITRACE_VERBOSE=2"
"${_test_library_path}")
set(_critical_trace_environment
"OMNITRACE_VERBOSE=2"
"OMNITRACE_USE_SAMPLING=OFF"
"OMNITRACE_USE_PROCESS_SAMPLING=OFF"
"OMNITRACE_CRITICAL_TRACE=ON"
"OMNITRACE_CRITICAL_TRACE_DEBUG=ON"
"OMNITRACE_TRACE_THREAD_LOCKS=ON"
"OMNITRACE_TRACE_THREAD_SPIN_LOCKS=ON"
"OMNITRACE_TRACE_THREAD_RW_LOCKS=ON"
"OMNITRACE_COUT_OUTPUT=ON"
"OMNITRACE_TIME_OUTPUT=OFF"
"OMNITRACE_TIMELINE_PROFILE=OFF"
"${_test_library_path}")
set(_ompt_environment
"OMNITRACE_TRACE=ON"
"OMNITRACE_PROFILE=ON"
"OMNITRACE_TIME_OUTPUT=OFF"
"OMNITRACE_USE_OMPT=ON"
"OMNITRACE_CRITICAL_TRACE=OFF"
"OMNITRACE_TIMEMORY_COMPONENTS=wall_clock,trip_count,peak_rss"
"${_test_openmp_env}"
"${_test_library_path}")
@@ -136,7 +120,7 @@ set(_timemory_environment
"${_test_openmp_env}"
"${_test_library_path}")
set(_test_environment ${_base_environment} "OMNITRACE_CRITICAL_TRACE=OFF")
set(_test_environment ${_base_environment})
set(_causal_environment
"${_test_openmp_env}" "${_test_library_path}" "OMNITRACE_TIME_OUTPUT=OFF"
@@ -159,7 +143,6 @@ set(_attach_environment
"OMNITRACE_PROFILE=ON"
"OMNITRACE_USE_SAMPLING=OFF"
"OMNITRACE_USE_PROCESS_SAMPLING=ON"
"OMNITRACE_USE_CRITICAL_TRACE=OFF"
"OMNITRACE_USE_OMPT=ON"
"OMNITRACE_USE_KOKKOSP=ON"
"OMNITRACE_TIME_OUTPUT=OFF"
tests/omnitrace-user-api-tests.cmake
+1 -1
View File
@@ -23,7 +23,7 @@ omnitrace_add_test(
return
args
RUN_ARGS 10 ${NUM_THREADS} 1000
ENVIRONMENT "${_base_environment};OMNITRACE_CRITICAL_TRACE=OFF"
ENVIRONMENT "${_base_environment}"
REWRITE_RUN_PASS_REGEX "Pushing custom region :: run.10. x 1000"
RUNTIME_PASS_REGEX "Pushing custom region :: run.10. x 1000"
SAMPLING_PASS_REGEX "Pushing custom region :: run.10. x 1000"