Relaxed module/function restrictions (#33)

* Relaxed module/function restrictions

* Updated tests

CMakeLists.txt

@@ -67,6 +67,7 @@ omnitrace_add_option(CMAKE_INSTALL_RPATH_USE_LINK_PATH "Enable rpath to linked l
 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_PAPI "Enable HW counter support via PAPI" OFF)
 omnitrace_add_option(OMNITRACE_USE_ROCTRACER "Enable roctracer support"
                      ${OMNITRACE_USE_HIP})
 omnitrace_add_option(
@@ -98,6 +99,9 @@ if(NOT OMNITRACE_USE_HIP)
     set(OMNITRACE_USE_ROCTRACER
         OFF
         CACHE BOOL "Disabled via OMNITRACE_USE_HIP=OFF" FORCE)
+    set(OMNITRACE_USE_ROCM_SMI
+        OFF
+        CACHE BOOL "Disabled via OMNITRACE_USE_HIP=OFF" FORCE)
 endif()
 
 if(OMNITRACE_BUILD_TESTING)
@@ -143,6 +147,7 @@ if(OMNITRACE_USE_HIP OR OMNITRACE_USE_ROCTRACER)
     if(HIP_VERSION_MAJOR GREATER_EQUAL 4 AND HIP_VERSION_MINOR GREATER 3)
         set(roctracer_kfdwrapper_LIBRARY)
     endif()
+    omnitrace_add_feature(HIP_VERSION "HIP version")
 else()
     set(HIP_VERSION "0.0.0")
     set(HIP_VERSION_MAJOR 0)

cmake/Packages.cmake

@@ -356,6 +356,9 @@ set(TIMEMORY_USE_PERFETTO
 set(TIMEMORY_USE_OMPT
     ${OMNITRACE_USE_OMPT}
     CACHE BOOL "Enable OMPT support in timemory" FORCE)
+set(TIMEMORY_USE_PAPI
+    ${OMNITRACE_USE_PAPI}
+    CACHE BOOL "Enable PAPI support in timemory" FORCE)
 set(TIMEMORY_USE_LIBUNWIND
     ON
     CACHE BOOL "Enable libunwind support in timemory")

source/bin/omnitrace/details.cpp

@@ -40,7 +40,8 @@ get_whole_function_names()
         "backtrace", "backtrace_symbols", "backtrace_symbols_fd", "sigaddset",
         "sigandset", "sigdelset", "sigemptyset", "sigfillset", "sighold", "sigisemptyset",
         "sigismember", "sigorset", "sigrelse", "sigvec", "strtok", "strstr", "sbrk",
-        "strxfrm", "atexit", "ompt_start_tool", "nanosleep",
+        "strxfrm", "atexit", "ompt_start_tool", "nanosleep", "cfree", "tolower",
+        "toupper", "fileno", "fileno_unlocked",
         // below are functions which never terminate
         "rocr::core::Signal::WaitAny", "rocr::core::Runtime::AsyncEventsLoop",
         "rocr::core::BusyWaitSignal::WaitAcquire",

source/bin/omnitrace/omnitrace.cpp

@@ -1882,9 +1882,8 @@ instrument_module(const string_t& file_name)
         "^(malloc|(f|)lock|sig|sem)[a-z_]+(|64|_r|_l)\\.c$"
     };
     static std::regex core_lib_regex{
-        "^(lib|)(c|z|rt|dl|dw|util|zstd|elf|pthread|open[\\-]rte|open[\\-]pal|"
-        "gcc_s|tcmalloc|profiler|tbbmalloc|tbbmalloc_proxy|event_pthreads|ltdl|"
-        "stdc\\+\\+|malloc|selinux|pcre[0-9]+)(-|\\.)",
+        "^(lib|)(c|dl|dw|pthread|tcmalloc|profiler|"
+        "tbbmalloc|tbbmalloc_proxy|malloc|stdc\\+\\+)(-|\\.)",
         regex_opts
     };
     static std::regex prefix_regex{ "^(_|\\.[a-zA-Z0-9])", regex_opts };
@@ -1950,22 +1949,16 @@ instrument_entity(const string_t& function_name)
     };
 
     static std::regex exclude(
-        "(omnitrace|tim::|N3tim|MPI_Init|MPI_Finalize|::__[A-Za-z]|"
-        "dyninst|tm_clones|malloc$|calloc$|free$|realloc$|std::addressof)",
+        "(omnitrace|tim::|N3tim|MPI_Init|MPI_Finalize|dyninst|tm_clones)", regex_opts);
+    static std::regex exclude_cxx(
+        "(std::_Sp_counted_base|std::(use|has)_facet|std::locale|::sentry|^std::_|::_(M|"
+        "S)_|::basic_string[a-zA-Z,<>: ]+::_M_create)",
         regex_opts);
-    static std::regex exclude_cxx("(std::_Sp_counted_base|std::use_facet)", regex_opts);
-    static std::regex leading(
-        "^(_|\\.|frame_dummy|\\(|targ|new|delete|operator new|operator delete|"
-        "std::allocat|nvtx|gcov|TAU|tau|Tau|dyn|RT|sys|pthread|posix|clone|"
-        "virtual thunk|non-virtual thunk|transaction clone|"
-        "RtsLayer|DYNINST|PthreadLayer|threaded_func|PMPI|"
-        "Kokkos::Impl::|Kokkos::Experimental::Impl::|Kokkos::impl_|"
-        "Kokkos::[A-Za-z]+::impl_|Kokkos::Tools::|Kokkos::Profiling::|"
-        "kmp_threadprivate_)",
-        regex_opts);
-    static std::regex trailing("(\\.part\\.[0-9]+|\\.constprop\\.[0-9]+|\\.|\\.[0-9]+)$",
-                               regex_opts);
-    static strset_t   whole = []() {
+    static std::regex leading("^(_|\\.|frame_dummy|\\(|targ|kmp_threadprivate_)",
+                              regex_opts);
+    static std::regex trailing(
+        "(_|\\.part\\.[0-9]+|\\.constprop\\.[0-9]+|\\.|\\.[0-9]+)$", regex_opts);
+    static strset_t whole = []() {
         auto _v   = get_whole_function_names();
         auto _ret = _v;
         for(std::string _ext : { "64", "_l", "_r" })

tests/CMakeLists.txt

@@ -243,7 +243,6 @@ omnitrace_add_test(
     NUM_PROCS ${NUM_PROCS}
     REWRITE_ARGS -e -v 2
     RUNTIME_ARGS -e -v 1 --label file line return args
-    RUN_ARGS ""
     ENVIRONMENT "${_base_environment};OMNITRACE_CRITICAL_TRACE=ON")
 
 omnitrace_add_test(
@@ -264,7 +263,6 @@ omnitrace_add_test(
         --dyninst-options
         DelayedParsing
         TypeChecking
-    RUN_ARGS ""
     ENVIRONMENT "${_fast_environment}")
 
 omnitrace_add_test(
@@ -287,12 +285,26 @@ omnitrace_add_test(
 omnitrace_add_test(
     NAME user-api
     TARGET user-api
-    REWRITE_ARGS -e -v 2 --min-address-range-loop=64
+    REWRITE_ARGS
+        -e
+        -v
+        2
+        -l
+        --min-address-range=0
+        --min-address-range-loop=0
+        --min-instructions=8
+        -E
+        custom_push_region
     RUNTIME_ARGS
         -e
         -v
         1
-        --min-address-range-loop=64
+        -l
+        --min-address-range=0
+        --min-address-range-loop=0
+        --min-instructions=8
+        -E
+        custom_push_region
         --label
         file
         line