diff --git a/openmp/libomptarget/DeviceRTL/include/Debug.h b/openmp/libomptarget/DeviceRTL/include/Debug.h
--- a/openmp/libomptarget/DeviceRTL/include/Debug.h
+++ b/openmp/libomptarget/DeviceRTL/include/Debug.h
@@ -49,13 +49,13 @@
 /// Enter a debugging scope for performing function traces. Enabled with
 /// FunctionTracting set in the debug kind.
 #define FunctionTracingRAII()                                                  \
-  DebugEntryRAII Entry(__LINE__, __PRETTY_FUNCTION__);
+  DebugEntryRAII Entry(__FILE__, __LINE__, __PRETTY_FUNCTION__);
 
 /// An RAII class for handling entries to debug locations. The current location
 /// and function will be printed on entry. Nested levels increase the
 /// indentation shown in the debugging output.
 struct DebugEntryRAII {
-  DebugEntryRAII(const unsigned Line, const char *Function);
+  DebugEntryRAII(const char *File, const unsigned Line, const char *Function);
   ~DebugEntryRAII();
 };
 
diff --git a/openmp/libomptarget/DeviceRTL/src/Debug.cpp b/openmp/libomptarget/DeviceRTL/src/Debug.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Debug.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Debug.cpp
@@ -12,6 +12,7 @@
 
 #include "Debug.h"
 #include "Configuration.h"
+#include "Interface.h"
 #include "Mapping.h"
 #include "Types.h"
 
@@ -41,14 +42,15 @@
 static uint32_t Level = 0;
 #pragma omp allocate(Level) allocator(omp_pteam_mem_alloc)
 
-DebugEntryRAII::DebugEntryRAII(const unsigned Line, const char *Function) {
+DebugEntryRAII::DebugEntryRAII(const char *File, const unsigned Line,
+                               const char *Function) {
   if (config::isDebugMode(config::DebugKind::FunctionTracing) &&
-      mapping::getThreadIdInBlock() == 0) {
+      mapping::getThreadIdInBlock() == 0 && mapping::getBlockId() == 0) {
 
     for (int I = 0; I < Level; ++I)
       PRINTF("%s", "  ");
 
-    PRINTF("Line %u: Thread %u Entering %s:%u\n", Line,
+    PRINTF("%s:%u: Thread %u Entering %s\n", File, Line,
            mapping::getThreadIdInBlock(), Function);
     Level++;
   }
@@ -56,7 +58,7 @@
 
 DebugEntryRAII::~DebugEntryRAII() {
   if (config::isDebugMode(config::DebugKind::FunctionTracing) &&
-      mapping::getThreadIdInBlock() == 0)
+      mapping::getThreadIdInBlock() == 0 && mapping::getBlockId() == 0)
     Level--;
 }
 
diff --git a/openmp/libomptarget/DeviceRTL/src/Kernel.cpp b/openmp/libomptarget/DeviceRTL/src/Kernel.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Kernel.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Kernel.cpp
@@ -30,6 +30,7 @@
 
 /// Simple generic state machine for worker threads.
 static void genericStateMachine(IdentTy *Ident) {
+  FunctionTracingRAII();
 
   uint32_t TId = mapping::getThreadIdInBlock();
 
@@ -66,6 +67,7 @@
 ///
 int32_t __kmpc_target_init(IdentTy *Ident, int8_t Mode,
                            bool UseGenericStateMachine, bool) {
+  FunctionTracingRAII();
   const bool IsSPMD = Mode & OMP_TGT_EXEC_MODE_SPMD;
   if (IsSPMD) {
     inititializeRuntime(/* IsSPMD */ true);
@@ -98,6 +100,7 @@
 /// \param Ident Source location identification, can be NULL.
 ///
 void __kmpc_target_deinit(IdentTy *Ident, int8_t Mode, bool) {
+  FunctionTracingRAII();
   const bool IsSPMD = Mode & OMP_TGT_EXEC_MODE_SPMD;
   state::assumeInitialState(IsSPMD);
   if (IsSPMD)
@@ -107,7 +110,10 @@
   state::ParallelRegionFn = nullptr;
 }
 
-int8_t __kmpc_is_spmd_exec_mode() { return mapping::isSPMDMode(); }
+int8_t __kmpc_is_spmd_exec_mode() {
+  FunctionTracingRAII();
+  return mapping::isSPMDMode();
+}
 }
 
 #pragma omp end declare target
diff --git a/openmp/libomptarget/DeviceRTL/src/Mapping.cpp b/openmp/libomptarget/DeviceRTL/src/Mapping.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Mapping.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Mapping.cpp
@@ -231,10 +231,12 @@
 
 extern "C" {
 __attribute__((noinline)) uint32_t __kmpc_get_hardware_thread_id_in_block() {
+  FunctionTracingRAII();
   return mapping::getThreadIdInBlock();
 }
 
 __attribute__((noinline)) uint32_t __kmpc_get_hardware_num_threads_in_block() {
+  FunctionTracingRAII();
   return mapping::getNumberOfProcessorElements();
 }
 }
diff --git a/openmp/libomptarget/DeviceRTL/src/Misc.cpp b/openmp/libomptarget/DeviceRTL/src/Misc.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Misc.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Misc.cpp
@@ -11,6 +11,8 @@
 
 #include "Types.h"
 
+#include "Debug.h"
+
 #pragma omp declare target
 
 namespace _OMP {
@@ -60,9 +62,15 @@
 ///{
 
 extern "C" {
-int32_t __kmpc_cancellationpoint(IdentTy *, int32_t, int32_t) { return 0; }
+int32_t __kmpc_cancellationpoint(IdentTy *, int32_t, int32_t) {
+  FunctionTracingRAII();
+  return 0;
+}
 
-int32_t __kmpc_cancel(IdentTy *, int32_t, int32_t) { return 0; }
+int32_t __kmpc_cancel(IdentTy *, int32_t, int32_t) {
+  FunctionTracingRAII();
+  return 0;
+}
 
 double omp_get_wtick(void) { return _OMP::impl::getWTick(); }
 
diff --git a/openmp/libomptarget/DeviceRTL/src/Parallelism.cpp b/openmp/libomptarget/DeviceRTL/src/Parallelism.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Parallelism.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Parallelism.cpp
@@ -66,6 +66,7 @@
 // Invoke an outlined parallel function unwrapping arguments (up to 32).
 void invokeMicrotask(int32_t global_tid, int32_t bound_tid, void *fn,
                      void **args, int64_t nargs) {
+  DebugEntryRAII Entry(__FILE__, __LINE__, "<OpenMP Outlined Function>");
   switch (nargs) {
 #include "generated_microtask_cases.gen"
   default:
@@ -81,6 +82,7 @@
 void __kmpc_parallel_51(IdentTy *ident, int32_t, int32_t if_expr,
                         int32_t num_threads, int proc_bind, void *fn,
                         void *wrapper_fn, void **args, int64_t nargs) {
+  FunctionTracingRAII();
 
   uint32_t TId = mapping::getThreadIdInBlock();
   // Handle the serialized case first, same for SPMD/non-SPMD.
@@ -149,6 +151,7 @@
 
 __attribute__((noinline)) bool
 __kmpc_kernel_parallel(ParallelRegionFnTy *WorkFn) {
+  FunctionTracingRAII();
   // Work function and arguments for L1 parallel region.
   *WorkFn = state::ParallelRegionFn;
 
@@ -163,6 +166,7 @@
 }
 
 __attribute__((noinline)) void __kmpc_kernel_end_parallel() {
+  FunctionTracingRAII();
   // In case we have modified an ICV for this thread before a ThreadState was
   // created. We drop it now to not contaminate the next parallel region.
   ASSERT(!mapping::isSPMDMode());
@@ -172,27 +176,40 @@
 }
 
 void __kmpc_serialized_parallel(IdentTy *, uint32_t TId) {
+  FunctionTracingRAII();
   state::enterDataEnvironment();
   ++icv::Level;
 }
 
 void __kmpc_end_serialized_parallel(IdentTy *, uint32_t TId) {
+  FunctionTracingRAII();
   state::exitDataEnvironment();
   --icv::Level;
 }
 
-uint16_t __kmpc_parallel_level(IdentTy *, uint32_t) { return omp_get_level(); }
+uint16_t __kmpc_parallel_level(IdentTy *, uint32_t) {
+  FunctionTracingRAII();
+  return omp_get_level();
+}
 
-int32_t __kmpc_global_thread_num(IdentTy *) { return omp_get_thread_num(); }
+int32_t __kmpc_global_thread_num(IdentTy *) {
+  FunctionTracingRAII();
+  return omp_get_thread_num();
+}
 
 void __kmpc_push_num_threads(IdentTy *, int32_t, int32_t NumThreads) {
+  FunctionTracingRAII();
   icv::NThreads = NumThreads;
 }
 
 void __kmpc_push_num_teams(IdentTy *loc, int32_t tid, int32_t num_teams,
-                           int32_t thread_limit) {}
+                           int32_t thread_limit) {
+  FunctionTracingRAII();
+}
 
-void __kmpc_push_proc_bind(IdentTy *loc, uint32_t tid, int proc_bind) {}
+void __kmpc_push_proc_bind(IdentTy *loc, uint32_t tid, int proc_bind) {
+  FunctionTracingRAII();
+}
 }
 
 #pragma omp end declare target
diff --git a/openmp/libomptarget/DeviceRTL/src/Reduction.cpp b/openmp/libomptarget/DeviceRTL/src/Reduction.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Reduction.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Reduction.cpp
@@ -176,6 +176,7 @@
 int32_t __kmpc_nvptx_parallel_reduce_nowait_v2(
     IdentTy *Loc, int32_t TId, int32_t num_vars, uint64_t reduce_size,
     void *reduce_data, ShuffleReductFnTy shflFct, InterWarpCopyFnTy cpyFct) {
+  FunctionTracingRAII();
   return nvptx_parallel_reduce_nowait(TId, num_vars, reduce_size, reduce_data,
                                       shflFct, cpyFct, mapping::isSPMDMode(),
                                       false);
@@ -186,6 +187,7 @@
     void *reduce_data, ShuffleReductFnTy shflFct, InterWarpCopyFnTy cpyFct,
     ListGlobalFnTy lgcpyFct, ListGlobalFnTy lgredFct, ListGlobalFnTy glcpyFct,
     ListGlobalFnTy glredFct) {
+  FunctionTracingRAII();
 
   // Terminate all threads in non-SPMD mode except for the master thread.
   uint32_t ThreadId = mapping::getThreadIdInBlock();
@@ -310,9 +312,9 @@
   return 0;
 }
 
-void __kmpc_nvptx_end_reduce(int32_t TId) {}
+void __kmpc_nvptx_end_reduce(int32_t TId) { FunctionTracingRAII(); }
 
-void __kmpc_nvptx_end_reduce_nowait(int32_t TId) {}
+void __kmpc_nvptx_end_reduce_nowait(int32_t TId) { FunctionTracingRAII(); }
 }
 
 #pragma omp end declare target
diff --git a/openmp/libomptarget/DeviceRTL/src/State.cpp b/openmp/libomptarget/DeviceRTL/src/State.cpp
--- a/openmp/libomptarget/DeviceRTL/src/State.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/State.cpp
@@ -496,10 +496,12 @@
 
 extern "C" {
 __attribute__((noinline)) void *__kmpc_alloc_shared(uint64_t Bytes) {
+  FunctionTracingRAII();
   return memory::allocShared(Bytes, "Frontend alloc shared");
 }
 
 __attribute__((noinline)) void __kmpc_free_shared(void *Ptr, uint64_t Bytes) {
+  FunctionTracingRAII();
   memory::freeShared(Ptr, Bytes, "Frontend free shared");
 }
 
@@ -521,6 +523,7 @@
     allocator(omp_pteam_mem_alloc)
 
 void __kmpc_begin_sharing_variables(void ***GlobalArgs, uint64_t nArgs) {
+  FunctionTracingRAII();
   if (nArgs <= NUM_SHARED_VARIABLES_IN_SHARED_MEM) {
     SharedMemVariableSharingSpacePtr = &SharedMemVariableSharingSpace[0];
   } else {
@@ -531,11 +534,13 @@
 }
 
 void __kmpc_end_sharing_variables() {
+  FunctionTracingRAII();
   if (SharedMemVariableSharingSpacePtr != &SharedMemVariableSharingSpace[0])
     memory::freeGlobal(SharedMemVariableSharingSpacePtr, "new extended args");
 }
 
 void __kmpc_get_shared_variables(void ***GlobalArgs) {
+  FunctionTracingRAII();
   *GlobalArgs = SharedMemVariableSharingSpacePtr;
 }
 }
diff --git a/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp b/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Synchronization.cpp
@@ -254,16 +254,18 @@
 }
 
 extern "C" {
-void __kmpc_ordered(IdentTy *Loc, int32_t TId) {}
+void __kmpc_ordered(IdentTy *Loc, int32_t TId) { FunctionTracingRAII(); }
 
-void __kmpc_end_ordered(IdentTy *Loc, int32_t TId) {}
+void __kmpc_end_ordered(IdentTy *Loc, int32_t TId) { FunctionTracingRAII(); }
 
 int32_t __kmpc_cancel_barrier(IdentTy *Loc, int32_t TId) {
+  FunctionTracingRAII();
   __kmpc_barrier(Loc, TId);
   return 0;
 }
 
 void __kmpc_barrier(IdentTy *Loc, int32_t TId) {
+  FunctionTracingRAII();
   if (mapping::isMainThreadInGenericMode())
     return __kmpc_flush(Loc);
 
@@ -275,34 +277,49 @@
 
 __attribute__((noinline)) void __kmpc_barrier_simple_spmd(IdentTy *Loc,
                                                           int32_t TId) {
+  FunctionTracingRAII();
   synchronize::threads();
 }
 
 int32_t __kmpc_master(IdentTy *Loc, int32_t TId) {
+  FunctionTracingRAII();
   return omp_get_team_num() == 0;
 }
 
-void __kmpc_end_master(IdentTy *Loc, int32_t TId) {}
+void __kmpc_end_master(IdentTy *Loc, int32_t TId) { FunctionTracingRAII(); }
 
 int32_t __kmpc_single(IdentTy *Loc, int32_t TId) {
+  FunctionTracingRAII();
   return __kmpc_master(Loc, TId);
 }
 
 void __kmpc_end_single(IdentTy *Loc, int32_t TId) {
+  FunctionTracingRAII();
   // The barrier is explicitly called.
 }
 
-void __kmpc_flush(IdentTy *Loc) { fence::kernel(__ATOMIC_SEQ_CST); }
+void __kmpc_flush(IdentTy *Loc) {
+  FunctionTracingRAII();
+  fence::kernel(__ATOMIC_SEQ_CST);
+}
 
-uint64_t __kmpc_warp_active_thread_mask(void) { return mapping::activemask(); }
+uint64_t __kmpc_warp_active_thread_mask(void) {
+  FunctionTracingRAII();
+  return mapping::activemask();
+}
 
-void __kmpc_syncwarp(uint64_t Mask) { synchronize::warp(Mask); }
+void __kmpc_syncwarp(uint64_t Mask) {
+  FunctionTracingRAII();
+  synchronize::warp(Mask);
+}
 
 void __kmpc_critical(IdentTy *Loc, int32_t TId, CriticalNameTy *Name) {
+  FunctionTracingRAII();
   omp_set_lock(reinterpret_cast<omp_lock_t *>(Name));
 }
 
 void __kmpc_end_critical(IdentTy *Loc, int32_t TId, CriticalNameTy *Name) {
+  FunctionTracingRAII();
   omp_unset_lock(reinterpret_cast<omp_lock_t *>(Name));
 }
 
diff --git a/openmp/libomptarget/DeviceRTL/src/Tasking.cpp b/openmp/libomptarget/DeviceRTL/src/Tasking.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Tasking.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Tasking.cpp
@@ -26,6 +26,7 @@
                                         uint64_t TaskSizeInclPrivateValues,
                                         uint64_t SharedValuesSize,
                                         TaskFnTy TaskFn) {
+  FunctionTracingRAII();
   auto TaskSizeInclPrivateValuesPadded =
       utils::roundUp(TaskSizeInclPrivateValues, uint64_t(sizeof(void *)));
   auto TaskSizeTotal = TaskSizeInclPrivateValuesPadded + SharedValuesSize;
@@ -40,12 +41,14 @@
 
 int32_t __kmpc_omp_task(IdentTy *Loc, uint32_t TId,
                         TaskDescriptorTy *TaskDescriptor) {
+  FunctionTracingRAII();
   return __kmpc_omp_task_with_deps(Loc, TId, TaskDescriptor, 0, 0, 0, 0);
 }
 
 int32_t __kmpc_omp_task_with_deps(IdentTy *Loc, uint32_t TId,
                                   TaskDescriptorTy *TaskDescriptor, int32_t,
                                   void *, int32_t, void *) {
+  FunctionTracingRAII();
   state::DateEnvironmentRAII DERAII;
 
   TaskDescriptor->TaskFn(0, TaskDescriptor);
@@ -56,31 +59,42 @@
 
 void __kmpc_omp_task_begin_if0(IdentTy *Loc, uint32_t TId,
                                TaskDescriptorTy *TaskDescriptor) {
+  FunctionTracingRAII();
   state::enterDataEnvironment();
 }
 
 void __kmpc_omp_task_complete_if0(IdentTy *Loc, uint32_t TId,
                                   TaskDescriptorTy *TaskDescriptor) {
+  FunctionTracingRAII();
   state::exitDataEnvironment();
 
   memory::freeGlobal(TaskDescriptor, "explicit task descriptor");
 }
 
 void __kmpc_omp_wait_deps(IdentTy *Loc, uint32_t TId, int32_t, void *, int32_t,
-                          void *) {}
+                          void *) {
+  FunctionTracingRAII();
+}
 
-void __kmpc_taskgroup(IdentTy *Loc, uint32_t TId) {}
+void __kmpc_taskgroup(IdentTy *Loc, uint32_t TId) { FunctionTracingRAII(); }
 
-void __kmpc_end_taskgroup(IdentTy *Loc, uint32_t TId) {}
+void __kmpc_end_taskgroup(IdentTy *Loc, uint32_t TId) { FunctionTracingRAII(); }
 
-int32_t __kmpc_omp_taskyield(IdentTy *Loc, uint32_t TId, int) { return 0; }
+int32_t __kmpc_omp_taskyield(IdentTy *Loc, uint32_t TId, int) {
+  FunctionTracingRAII();
+  return 0;
+}
 
-int32_t __kmpc_omp_taskwait(IdentTy *Loc, uint32_t TId) { return 0; }
+int32_t __kmpc_omp_taskwait(IdentTy *Loc, uint32_t TId) {
+  FunctionTracingRAII();
+  return 0;
+}
 
 void __kmpc_taskloop(IdentTy *Loc, uint32_t TId,
                      TaskDescriptorTy *TaskDescriptor, int,
                      uint64_t *LowerBound, uint64_t *UpperBound, int64_t, int,
                      int32_t, uint64_t, void *) {
+  FunctionTracingRAII();
   // Skip task entirely if empty iteration space.
   if (*LowerBound > *UpperBound)
     return;
diff --git a/openmp/libomptarget/DeviceRTL/src/Utils.cpp b/openmp/libomptarget/DeviceRTL/src/Utils.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Utils.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Utils.cpp
@@ -11,6 +11,7 @@
 
 #include "Utils.h"
 
+#include "Debug.h"
 #include "Interface.h"
 #include "Mapping.h"
 
@@ -129,10 +130,12 @@
 
 extern "C" {
 int32_t __kmpc_shuffle_int32(int32_t Val, int16_t Delta, int16_t SrcLane) {
+  FunctionTracingRAII();
   return impl::shuffleDown(lanes::All, Val, Delta, SrcLane);
 }
 
 int64_t __kmpc_shuffle_int64(int64_t Val, int16_t Delta, int16_t Width) {
+  FunctionTracingRAII();
   uint32_t lo, hi;
   utils::unpack(Val, lo, hi);
   hi = impl::shuffleDown(lanes::All, hi, Delta, Width);
diff --git a/openmp/libomptarget/DeviceRTL/src/Workshare.cpp b/openmp/libomptarget/DeviceRTL/src/Workshare.cpp
--- a/openmp/libomptarget/DeviceRTL/src/Workshare.cpp
+++ b/openmp/libomptarget/DeviceRTL/src/Workshare.cpp
@@ -474,6 +474,7 @@
 // init
 void __kmpc_dispatch_init_4(IdentTy *loc, int32_t tid, int32_t schedule,
                             int32_t lb, int32_t ub, int32_t st, int32_t chunk) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = pushDST();
   omptarget_nvptx_LoopSupport<int32_t, int32_t>::dispatch_init(
       loc, tid, (kmp_sched_t)schedule, lb, ub, st, chunk, DST);
@@ -482,6 +483,7 @@
 void __kmpc_dispatch_init_4u(IdentTy *loc, int32_t tid, int32_t schedule,
                              uint32_t lb, uint32_t ub, int32_t st,
                              int32_t chunk) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = pushDST();
   omptarget_nvptx_LoopSupport<uint32_t, int32_t>::dispatch_init(
       loc, tid, (kmp_sched_t)schedule, lb, ub, st, chunk, DST);
@@ -489,6 +491,7 @@
 
 void __kmpc_dispatch_init_8(IdentTy *loc, int32_t tid, int32_t schedule,
                             int64_t lb, int64_t ub, int64_t st, int64_t chunk) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = pushDST();
   omptarget_nvptx_LoopSupport<int64_t, int64_t>::dispatch_init(
       loc, tid, (kmp_sched_t)schedule, lb, ub, st, chunk, DST);
@@ -497,6 +500,7 @@
 void __kmpc_dispatch_init_8u(IdentTy *loc, int32_t tid, int32_t schedule,
                              uint64_t lb, uint64_t ub, int64_t st,
                              int64_t chunk) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = pushDST();
   omptarget_nvptx_LoopSupport<uint64_t, int64_t>::dispatch_init(
       loc, tid, (kmp_sched_t)schedule, lb, ub, st, chunk, DST);
@@ -505,6 +509,7 @@
 // next
 int __kmpc_dispatch_next_4(IdentTy *loc, int32_t tid, int32_t *p_last,
                            int32_t *p_lb, int32_t *p_ub, int32_t *p_st) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = peekDST();
   return omptarget_nvptx_LoopSupport<int32_t, int32_t>::dispatch_next(
       loc, tid, p_last, p_lb, p_ub, p_st, DST);
@@ -512,6 +517,7 @@
 
 int __kmpc_dispatch_next_4u(IdentTy *loc, int32_t tid, int32_t *p_last,
                             uint32_t *p_lb, uint32_t *p_ub, int32_t *p_st) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = peekDST();
   return omptarget_nvptx_LoopSupport<uint32_t, int32_t>::dispatch_next(
       loc, tid, p_last, p_lb, p_ub, p_st, DST);
@@ -519,6 +525,7 @@
 
 int __kmpc_dispatch_next_8(IdentTy *loc, int32_t tid, int32_t *p_last,
                            int64_t *p_lb, int64_t *p_ub, int64_t *p_st) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = peekDST();
   return omptarget_nvptx_LoopSupport<int64_t, int64_t>::dispatch_next(
       loc, tid, p_last, p_lb, p_ub, p_st, DST);
@@ -526,6 +533,7 @@
 
 int __kmpc_dispatch_next_8u(IdentTy *loc, int32_t tid, int32_t *p_last,
                             uint64_t *p_lb, uint64_t *p_ub, int64_t *p_st) {
+  FunctionTracingRAII();
   DynamicScheduleTracker *DST = peekDST();
   return omptarget_nvptx_LoopSupport<uint64_t, int64_t>::dispatch_next(
       loc, tid, p_last, p_lb, p_ub, p_st, DST);
@@ -533,21 +541,25 @@
 
 // fini
 void __kmpc_dispatch_fini_4(IdentTy *loc, int32_t tid) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<int32_t, int32_t>::dispatch_fini();
   popDST();
 }
 
 void __kmpc_dispatch_fini_4u(IdentTy *loc, int32_t tid) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<uint32_t, int32_t>::dispatch_fini();
   popDST();
 }
 
 void __kmpc_dispatch_fini_8(IdentTy *loc, int32_t tid) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<int64_t, int64_t>::dispatch_fini();
   popDST();
 }
 
 void __kmpc_dispatch_fini_8u(IdentTy *loc, int32_t tid) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<uint64_t, int64_t>::dispatch_fini();
   popDST();
 }
@@ -560,6 +572,7 @@
                               int32_t schedtype, int32_t *plastiter,
                               int32_t *plower, int32_t *pupper,
                               int32_t *pstride, int32_t incr, int32_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<int32_t, int32_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -569,6 +582,7 @@
                                int32_t schedtype, int32_t *plastiter,
                                uint32_t *plower, uint32_t *pupper,
                                int32_t *pstride, int32_t incr, int32_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<uint32_t, int32_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -578,6 +592,7 @@
                               int32_t schedtype, int32_t *plastiter,
                               int64_t *plower, int64_t *pupper,
                               int64_t *pstride, int64_t incr, int64_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<int64_t, int64_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -587,6 +602,7 @@
                                int32_t schedtype, int32_t *plastiter,
                                uint64_t *plower, uint64_t *pupper,
                                int64_t *pstride, int64_t incr, int64_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<uint64_t, int64_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -597,6 +613,7 @@
                                      int32_t *plower, int32_t *pupper,
                                      int32_t *pstride, int32_t incr,
                                      int32_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<int32_t, int32_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -607,6 +624,7 @@
                                       uint32_t *plower, uint32_t *pupper,
                                       int32_t *pstride, int32_t incr,
                                       int32_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<uint32_t, int32_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -617,6 +635,7 @@
                                      int64_t *plower, int64_t *pupper,
                                      int64_t *pstride, int64_t incr,
                                      int64_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<int64_t, int64_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
@@ -627,14 +646,19 @@
                                       uint64_t *plower, uint64_t *pupper,
                                       int64_t *pstride, int64_t incr,
                                       int64_t chunk) {
+  FunctionTracingRAII();
   omptarget_nvptx_LoopSupport<uint64_t, int64_t>::for_static_init(
       global_tid, schedtype, plastiter, plower, pupper, pstride, chunk,
       mapping::isSPMDMode());
 }
 
-void __kmpc_for_static_fini(IdentTy *loc, int32_t global_tid) {}
+void __kmpc_for_static_fini(IdentTy *loc, int32_t global_tid) {
+  FunctionTracingRAII();
+}
 
-void __kmpc_distribute_static_fini(IdentTy *loc, int32_t global_tid) {}
+void __kmpc_distribute_static_fini(IdentTy *loc, int32_t global_tid) {
+  FunctionTracingRAII();
+}
 }
 
 #pragma omp end declare target