[OMPT] Fix duplicate implicit_task_end events for master thread with GCC

[Hahnfeld]

Summary

For non-serialized parallel regions the master thread issued two callbacks: The first one in kmp_gsupport.c and the second in __kmp_join_call. Therefore only trigger the callback in kmp_gsupport.c for serialized parallel regions.

[Hahnfeld]

Ping

[omalyshe]

LGTM.

runtime/src/kmp_gsupport.c

xexpand(KMP_API_NAME_GOMP_PARALLEL_END)(void)
     thr = __kmp_threads[gtid];
 
     MKLOC(loc, "GOMP_parallel_end");
     KA_TRACE(20, ("GOMP_parallel_end: T#%d\n", gtid));
 
 
 #if OMPT_SUPPORT
     ompt_parallel_id_t parallel_id;
+    ompt_task_id_t serialized_task_id;
     ompt_frame_t *ompt_frame = NULL;
 
     if (ompt_enabled) {
         ompt_team_info_t *team_info = __ompt_get_teaminfo(0, NULL);
         parallel_id = team_info->parallel_id;
 
+        ompt_task_info_t *task_info = __ompt_get_taskinfo(0);
+        serialized_task_id = task_info->task_id;
+
         // Record that we re-entered the runtime system in the implicit
         // task frame representing the parallel region. 
-        ompt_frame = __ompt_get_task_frame_internal(0);
+        ompt_frame = &task_info->frame;
         ompt_frame->reenter_runtime_frame = __builtin_frame_address(0);
 
-#if OMPT_TRACE
-        if (ompt_enabled &&
-            ompt_callbacks.ompt_callback(ompt_event_implicit_task_end)) {
-            ompt_task_info_t *task_info = __ompt_get_taskinfo(0);
-            ompt_callbacks.ompt_callback(ompt_event_implicit_task_end)(
-                parallel_id, task_info->task_id);
-        }
-#endif
-
         // unlink if necessary. no-op if there is not a lightweight task.
         ompt_lw_taskteam_t *lwt = __ompt_lw_taskteam_unlink(thr);
         // GOMP allocates/frees lwt since it can't be kept on the stack
         if (lwt) { 
            __kmp_free(lwt);
      
 #if OMPT_SUPPORT
            if (ompt_enabled) {
#endif
         );
 #if OMPT_SUPPORT
         if (ompt_enabled) {
           ompt_frame->reenter_runtime_frame = NULL;
         }
 #endif
     }
     else {
+#if OMPT_SUPPORT && OMPT_TRACE
+        if (ompt_enabled &&
+            ompt_callbacks.ompt_callback(ompt_event_implicit_task_end)) {
+            ompt_callbacks.ompt_callback(ompt_event_implicit_task_end)(
+                parallel_id, serialized_task_id);
+        }
+#endif
+
         __kmpc_end_serialized_parallel(&loc, gtid);
 
 #if OMPT_SUPPORT
         if (ompt_enabled) {
             // Record that we re-entered the runtime system in the frame that 
             // created the parallel region.
             ompt_frame->reenter_runtime_frame = __builtin_frame_address(0);
 

runtime/test/ompt/omp_parallel.c

int main()
   // CHECK-DAG: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // CHECK: {{^}}[[MASTER_ID]]: ompt_event_parallel_end: parallel_id=[[PARALLEL_ID]], task_id=[[PARENT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
 
 
   // THREADS: {{^}}[[MASTER_ID:[0-9]+]]: ompt_event_parallel_begin: parent_task_id={{[0-9]+}}, parent_task_frame=0x{{[0-f]+}}, parallel_id=[[PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=0x{{[0-f]+}}, invoker={{.*}}
 
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   return 0;
 }

runtime/test/ompt/omp_parallel_nested.c

int main()
 
 
   // THREADS: {{^}}[[MASTER_ID:[0-9]+]]: ompt_event_parallel_begin: parent_task_id={{[0-9]+}}, parent_task_frame=0x{{[0-f]+}}, parallel_id=[[PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=0x{{[0-f]+}}, invoker=[[PARALLEL_INVOKER:.+]]
 
   // nested parallel masters
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION:0x[0-f]+]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // nested parallel worker threads
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   return 0;
 }

runtime/test/ompt/omp_parallel_nested_serialized.c

int main()
 
   // THREADS: {{^}}[[MASTER_ID:[0-9]+]]: ompt_event_parallel_begin: parent_task_id={{[0-9]+}}, parent_task_frame=0x{{[0-f]+}}, parallel_id=[[PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=0x{{[0-f]+}}, invoker=[[PARALLEL_INVOKER:.+]]
 
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION:0x[0-f]+]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[MASTER_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   // THREADS: {{^}}[[THREAD_ID:[0-9]+]]: ompt_event_implicit_task_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_begin: parent_task_id=[[IMPLICIT_TASK_ID]], parent_task_frame=0x{{[0-f]+}}, parallel_id=[[NESTED_PARALLEL_ID:[0-9]+]], requested_team_size=4, parallel_function=[[NESTED_PARALLEL_FUNCTION]], invoker=[[PARALLEL_INVOKER]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_begin: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID:[0-9]+]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[NESTED_IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_parallel_end: parallel_id=[[NESTED_PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]], invoker=[[PARALLEL_INVOKER]]
+  // THREADS-NOT: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_begin: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_barrier_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
   // THREADS: {{^}}[[THREAD_ID]]: ompt_event_implicit_task_end: parallel_id=[[PARALLEL_ID]], task_id=[[IMPLICIT_TASK_ID]]
 
   return 0;
 }