Improve performance of #pragma omp single

[tlwilmar]

Summary

Performance of single was considerably worse than plain barrier.
Adding a check to the code before the atomic compare and store helped
considerably.

Patch by Terry Wilmarth.

[hbae]

LGTM.

openmp/trunk/runtime/src/kmp_runtime.c

__kmp_enter_single( int gtid, ident_t *id_ref, int push_ws )
     } else {
         kmp_int32 old_this = th->th.th_local.this_construct;
 
         ++th->th.th_local.this_construct;
         /* try to set team count to thread count--success means thread got the
            single block
         */
         /* TODO: Should this be acquire or release? */
+        if (team->t.t_construct == old_this) {
             status = KMP_COMPARE_AND_STORE_ACQ32(&team->t.t_construct, old_this,
                                                  th->th.th_local.this_construct);
+        }
 #if USE_ITT_BUILD
         if ( __itt_metadata_add_ptr && __kmp_forkjoin_frames_mode == 3 && KMP_MASTER_GTID(gtid) &&
 #if OMP_40_ENABLED
             th->th.th_teams_microtask == NULL &&
 #endif
             team->t.t_active_level == 1 )
         {   // Only report metadata by master of active team at level 1
             __kmp_itt_metadata_single( id_ref );