diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index a8dcb16c27ac6f41e8b9f534eec43c527d32fc8a..cb9c6f1a9569a92c07a55f5ec7d25c49978dc987 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -99,6 +99,8 @@ struct Qdisc {
 	 * for 32-bit kernel.
 	 */
 	struct rcu_head         rcu;
+	/* onwer will use 4 Bytes, and the space is enough.*/
+	int			owner;
 #endif
 	/*
 	 * For performance sake on SMP, we put highly modified fields at the end
diff --git a/net/core/dev.c b/net/core/dev.c
index a9aca8971a59ea83c840478462b6caf100d133af..f570d33b3bc7e223d226f3acf6ed84c20d718487 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3472,6 +3472,10 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
 		return rc;
 	}
 
+	if (unlikely(READ_ONCE(q->owner) == smp_processor_id())) {
+		kfree_skb(skb);
+		return NET_XMIT_DROP;
+	}
 	/*
 	 * Heuristic to force contended enqueues to serialize on a
 	 * separate lock before trying to get qdisc main lock.
@@ -3507,7 +3511,9 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
 		qdisc_run_end(q);
 		rc = NET_XMIT_SUCCESS;
 	} else {
+		WRITE_ONCE(q->owner, smp_processor_id());
 		rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK;
+		WRITE_ONCE(q->owner, -1);
 		if (qdisc_run_begin(q)) {
 			if (unlikely(contended)) {
 				spin_unlock(&q->busylock);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 64ae84bea587ad74328ca2ec660c4aaadea0a617..18b587cf9db88e0b5bf6d8e629b27e50204a2223 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -881,6 +881,7 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
 	sch->enqueue = ops->enqueue;
 	sch->dequeue = ops->dequeue;
 	sch->dev_queue = dev_queue;
+	sch->owner = -1;
 	dev_hold(dev);
 	refcount_set(&sch->refcnt, 1);