diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 3ed6bcc4be72..177f240d59d4 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -37,8 +37,15 @@ enum qdisc_state_t {
 	__QDISC_STATE_SCHED,
 	__QDISC_STATE_DEACTIVATED,
 	__QDISC_STATE_MISSED,
+	__QDISC_STATE_DRAINING,
 };
 
+#define QDISC_STATE_MISSED	BIT(__QDISC_STATE_MISSED)
+#define QDISC_STATE_DRAINING	BIT(__QDISC_STATE_DRAINING)
+
+#define QDISC_STATE_NON_EMPTY	(QDISC_STATE_MISSED | \
+					QDISC_STATE_DRAINING)
+
 struct qdisc_size_table {
 	struct rcu_head		rcu;
 	struct list_head	list;
@@ -145,6 +152,11 @@ static inline bool qdisc_is_running(struct Qdisc *qdisc)
 	return (raw_read_seqcount(&qdisc->running) & 1) ? true : false;
 }
 
+static inline bool nolock_qdisc_is_empty(const struct Qdisc *qdisc)
+{
+	return !(READ_ONCE(qdisc->state) & QDISC_STATE_NON_EMPTY);
+}
+
 static inline bool qdisc_is_percpu_stats(const struct Qdisc *q)
 {
 	return q->flags & TCQ_F_CPUSTATS;
@@ -206,10 +218,8 @@ static inline void qdisc_run_end(struct Qdisc *qdisc)
 		spin_unlock(&qdisc->seqlock);
 
 		if (unlikely(test_bit(__QDISC_STATE_MISSED,
-				      &qdisc->state))) {
-			clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
+				      &qdisc->state)))
 			__netif_schedule(qdisc);
-		}
 	} else {
 		write_seqcount_end(&qdisc->running);
 	}
diff --git a/net/core/dev.c b/net/core/dev.c
index 50531a2d0b20..991d09b67bd9 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3852,10 +3852,33 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
 	qdisc_calculate_pkt_len(skb, q);
 
 	if (q->flags & TCQ_F_NOLOCK) {
-		rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK;
-		if (likely(!netif_xmit_frozen_or_stopped(txq)))
-			qdisc_run(q);
+		if (q->flags & TCQ_F_CAN_BYPASS && nolock_qdisc_is_empty(q) &&
+		    qdisc_run_begin(q)) {
+			/* Retest nolock_qdisc_is_empty() within the protection
+			 * of q->seqlock to protect from racing with requeuing.
+			 */
+			if (unlikely(!nolock_qdisc_is_empty(q))) {
+				rc = q->enqueue(skb, q, &to_free) &
+					NET_XMIT_MASK;
+				__qdisc_run(q);
+				qdisc_run_end(q);
 
+				goto no_lock_out;
+			}
+
+			qdisc_bstats_cpu_update(q, skb);
+			if (sch_direct_xmit(skb, q, dev, txq, NULL, true) &&
+			    !nolock_qdisc_is_empty(q))
+				__qdisc_run(q);
+
+			qdisc_run_end(q);
+			return NET_XMIT_SUCCESS;
+		}
+
+		rc = q->enqueue(skb, q, &to_free) & NET_XMIT_MASK;
+		qdisc_run(q);
+
+no_lock_out:
 		if (unlikely(to_free))
 			kfree_skb_list(to_free);
 		return rc;
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index e9c0afc8becc..9984ccc45946 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -52,6 +52,8 @@ static void qdisc_maybe_clear_missed(struct Qdisc *q,
 	 */
 	if (!netif_xmit_frozen_or_stopped(txq))
 		set_bit(__QDISC_STATE_MISSED, &q->state);
+	else
+		set_bit(__QDISC_STATE_DRAINING, &q->state);
 }
 
 /* Main transmission queue. */
@@ -164,9 +166,13 @@ static inline void dev_requeue_skb(struct sk_buff *skb, struct Qdisc *q)
 
 		skb = next;
 	}
-	if (lock)
+
+	if (lock) {
 		spin_unlock(lock);
-	__netif_schedule(q);
+		set_bit(__QDISC_STATE_MISSED, &q->state);
+	} else {
+		__netif_schedule(q);
+	}
 }
 
 static void try_bulk_dequeue_skb(struct Qdisc *q,
@@ -409,7 +415,11 @@ void __qdisc_run(struct Qdisc *q)
 	while (qdisc_restart(q, &packets)) {
 		quota -= packets;
 		if (quota <= 0) {
-			__netif_schedule(q);
+			if (q->flags & TCQ_F_NOLOCK)
+				set_bit(__QDISC_STATE_MISSED, &q->state);
+			else
+				__netif_schedule(q);
+
 			break;
 		}
 	}
@@ -698,13 +708,14 @@ retry:
 	if (likely(skb)) {
 		qdisc_update_stats_at_dequeue(qdisc, skb);
 	} else if (need_retry &&
-		   test_bit(__QDISC_STATE_MISSED, &qdisc->state)) {
+		   READ_ONCE(qdisc->state) & QDISC_STATE_NON_EMPTY) {
 		/* Delay clearing the STATE_MISSED here to reduce
 		 * the overhead of the second spin_trylock() in
 		 * qdisc_run_begin() and __netif_schedule() calling
 		 * in qdisc_run_end().
 		 */
 		clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
+		clear_bit(__QDISC_STATE_DRAINING, &qdisc->state);
 
 		/* Make sure dequeuing happens after clearing
 		 * STATE_MISSED.
@@ -1222,6 +1233,7 @@ static void dev_reset_queue(struct net_device *dev,
 	spin_unlock_bh(qdisc_lock(qdisc));
 	if (nolock) {
 		clear_bit(__QDISC_STATE_MISSED, &qdisc->state);
+		clear_bit(__QDISC_STATE_DRAINING, &qdisc->state);
 		spin_unlock_bh(&qdisc->seqlock);
 	}
 }