diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3529bf61826b..25c7c7e09cbd 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1567,6 +1567,13 @@ static bool load_too_imbalanced(long src_load, long dst_load,
 	return (imb > old_imb);
 }
 
+/*
+ * Maximum NUMA importance can be 1998 (2*999);
+ * SMALLIMP @ 30 would be close to 1998/64.
+ * Used to deter task migration.
+ */
+#define SMALLIMP	30
+
 /*
  * This checks if the overall compute and NUMA accesses of the system would
  * be improved if the source tasks was migrated to the target dst_cpu taking
@@ -1600,7 +1607,7 @@ static void task_numa_compare(struct task_numa_env *env,
 		goto unlock;
 
 	if (!cur) {
-		if (maymove || imp > env->best_imp)
+		if (maymove && moveimp >= env->best_imp)
 			goto assign;
 		else
 			goto unlock;
@@ -1643,15 +1650,21 @@ static void task_numa_compare(struct task_numa_env *env,
 			       task_weight(cur, env->dst_nid, dist);
 	}
 
-	if (imp <= env->best_imp)
-		goto unlock;
-
 	if (maymove && moveimp > imp && moveimp > env->best_imp) {
-		imp = moveimp - 1;
+		imp = moveimp;
 		cur = NULL;
 		goto assign;
 	}
 
+	/*
+	 * If the NUMA importance is less than SMALLIMP,
+	 * task migration might only result in ping pong
+	 * of tasks and also hurt performance due to cache
+	 * misses.
+	 */
+	if (imp < SMALLIMP || imp <= env->best_imp + SMALLIMP / 2)
+		goto unlock;
+
 	/*
 	 * In the overloaded case, try and keep the load balanced.
 	 */