[ITTNOTIFY] Correct barrier imbalance time in case of tasks

ittnotify fix for barrier imbalance time in case tasks exist. In the current
implementation, task execution time is included into aggregated time on a
barrier. This fix calculates task execution time and corrects the arrive time
by subtracting the task execution time.

Since __kmp_invoke_task() can not only be called on a barrier, the field
th.th_bar_arrive_time is used to check if the function was called at the
barrier (th.th_bar_arrive_time != 0). So for this check, th_bar_arrive_time
is set to zero right after the value is used on the barrier.

Differential Revision: http://reviews.llvm.org/D19030

llvm-svn: 266332

diff --git a/openmp/runtime/src/kmp_barrier.cpp b/openmp/runtime/src/kmp_barrier.cpp
index e449a30..6fef085 100644 (file)
--- a/openmp/runtime/src/kmp_barrier.cpp
+++ b/openmp/runtime/src/kmp_barrier.cpp
@@ -1198,8 +1198,11 @@ __kmp_barrier(enum barrier_type bt, int gtid, int is_split, size_t reduce_size,
                     if( __itt_metadata_add_ptr ) {
                         // Initialize with master's wait time
                         kmp_uint64 delta = cur_time - this_thr->th.th_bar_arrive_time;
+                        // Set arrive time to zero to be able to check it in __kmp_invoke_task(); the same is done inside the loop below
+                        this_thr->th.th_bar_arrive_time = 0;
                         for (i=1; i<nproc; ++i) {
                             delta += ( cur_time - other_threads[i]->th.th_bar_arrive_time );
+                            other_threads[i]->th.th_bar_arrive_time = 0;
                         }
                         __kmp_itt_metadata_imbalance(gtid, this_thr->th.th_frame_time, cur_time, delta, (kmp_uint64)( reduce != NULL));
                     }
@@ -1489,8 +1492,11 @@ __kmp_join_barrier(int gtid)
                 if( __itt_metadata_add_ptr ) {
                     // Initialize with master's wait time
                     kmp_uint64 delta = cur_time - this_thr->th.th_bar_arrive_time;
+                    // Set arrive time to zero to be able to check it in __kmp_invoke_task(); the same is done inside the loop below
+                    this_thr->th.th_bar_arrive_time = 0;
                     for (i=1; i<nproc; ++i) {
                         delta += ( cur_time - other_threads[i]->th.th_bar_arrive_time );
+                        other_threads[i]->th.th_bar_arrive_time = 0;
                     }
                     __kmp_itt_metadata_imbalance(gtid, this_thr->th.th_frame_time, cur_time, delta, 0);
                 }

diff --git a/openmp/runtime/src/kmp_tasking.c b/openmp/runtime/src/kmp_tasking.c
index 1122c49..044b7ff 100644 (file)
--- a/openmp/runtime/src/kmp_tasking.c
+++ b/openmp/runtime/src/kmp_tasking.c
@@ -1103,6 +1103,7 @@ static void
 __kmp_invoke_task( kmp_int32 gtid, kmp_task_t *task, kmp_taskdata_t * current_task )
 {
     kmp_taskdata_t * taskdata = KMP_TASK_TO_TASKDATA(task);
+    kmp_uint64 cur_time;
 #if OMP_40_ENABLED
     int discard = 0 /* false */;
 #endif
@@ -1126,6 +1127,13 @@ __kmp_invoke_task( kmp_int32 gtid, kmp_task_t *task, kmp_taskdata_t * current_ta
          }
 #endif
 
+#if USE_ITT_BUILD && USE_ITT_NOTIFY
+    if(__kmp_forkjoin_frames_mode == 3) {
+        // Get the current time stamp to measure task execution time to correct barrier imbalance time
+        cur_time = __itt_get_timestamp();
+    }
+#endif
+
 #if OMP_41_ENABLED
     // Proxy tasks are not handled by the runtime
     if ( taskdata->td_flags.proxy != TASK_PROXY )
@@ -1219,6 +1227,15 @@ __kmp_invoke_task( kmp_int32 gtid, kmp_task_t *task, kmp_taskdata_t * current_ta
 #endif
        __kmp_task_finish( gtid, task, current_task );
 
+#if USE_ITT_BUILD && USE_ITT_NOTIFY
+    // Barrier imbalance - correct arrive time after the task finished
+    if(__kmp_forkjoin_frames_mode == 3) {
+        kmp_info_t *this_thr = __kmp_threads [ gtid ];
+        if(this_thr->th.th_bar_arrive_time) {
+            this_thr->th.th_bar_arrive_time += (__itt_get_timestamp() - cur_time);
+        }
+    }
+#endif
     KA_TRACE(30, ("__kmp_invoke_task(exit): T#%d completed task %p, resuming task %p\n",
                   gtid, taskdata, current_task) );
     return;