diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index 1c78c7454c7c..a11afecd4482 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -718,6 +718,7 @@ enum {
 	TCA_FQ_CODEL_FLOWS,
 	TCA_FQ_CODEL_QUANTUM,
 	TCA_FQ_CODEL_CE_THRESHOLD,
+	TCA_FQ_CODEL_DROP_BATCH_SIZE,
 	__TCA_FQ_CODEL_MAX
 };
 
diff --git a/net/sched/sch_fq_codel.c b/net/sched/sch_fq_codel.c
index a5e420b3d4ab..e7b42b0d5145 100644
--- a/net/sched/sch_fq_codel.c
+++ b/net/sched/sch_fq_codel.c
@@ -59,6 +59,7 @@ struct fq_codel_sched_data {
 	u32		flows_cnt;	/* number of flows */
 	u32		perturbation;	/* hash perturbation */
 	u32		quantum;	/* psched_mtu(qdisc_dev(sch)); */
+	u32		drop_batch_size;
 	struct codel_params cparams;
 	struct codel_stats cstats;
 	u32		drop_overlimit;
@@ -135,17 +136,20 @@ static inline void flow_queue_add(struct fq_codel_flow *flow,
 	skb->next = NULL;
 }
 
-static unsigned int fq_codel_drop(struct Qdisc *sch)
+static unsigned int fq_codel_drop(struct Qdisc *sch, unsigned int max_packets)
 {
 	struct fq_codel_sched_data *q = qdisc_priv(sch);
 	struct sk_buff *skb;
 	unsigned int maxbacklog = 0, idx = 0, i, len;
 	struct fq_codel_flow *flow;
+	unsigned int threshold;
 
-	/* Queue is full! Find the fat flow and drop packet from it.
+	/* Queue is full! Find the fat flow and drop packet(s) from it.
 	 * This might sound expensive, but with 1024 flows, we scan
 	 * 4KB of memory, and we dont need to handle a complex tree
 	 * in fast path (packet queue/enqueue) with many cache misses.
+	 * In stress mode, we'll try to drop 64 packets from the flow,
+	 * amortizing this linear lookup to one cache line per drop.
 	 */
 	for (i = 0; i < q->flows_cnt; i++) {
 		if (q->backlogs[i] > maxbacklog) {
@@ -153,15 +157,24 @@ static unsigned int fq_codel_drop(struct Qdisc *sch)
 			idx = i;
 		}
 	}
+
+	/* Our goal is to drop half of this fat flow backlog */
+	threshold = maxbacklog >> 1;
+
 	flow = &q->flows[idx];
-	skb = dequeue_head(flow);
-	len = qdisc_pkt_len(skb);
+	len = 0;
+	i = 0;
+	do {
+		skb = dequeue_head(flow);
+		len += qdisc_pkt_len(skb);
+		kfree_skb(skb);
+	} while (++i < max_packets && len < threshold);
+
+	flow->dropped += i;
 	q->backlogs[idx] -= len;
-	sch->q.qlen--;
-	qdisc_qstats_drop(sch);
-	qdisc_qstats_backlog_dec(sch, skb);
-	kfree_skb(skb);
-	flow->dropped++;
+	sch->qstats.drops += i;
+	sch->qstats.backlog -= len;
+	sch->q.qlen -= i;
 	return idx;
 }
 
@@ -170,14 +183,14 @@ static unsigned int fq_codel_qdisc_drop(struct Qdisc *sch)
 	unsigned int prev_backlog;
 
 	prev_backlog = sch->qstats.backlog;
-	fq_codel_drop(sch);
+	fq_codel_drop(sch, 1U);
 	return prev_backlog - sch->qstats.backlog;
 }
 
 static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
 	struct fq_codel_sched_data *q = qdisc_priv(sch);
-	unsigned int idx, prev_backlog;
+	unsigned int idx, prev_backlog, prev_qlen;
 	struct fq_codel_flow *flow;
 	int uninitialized_var(ret);
 
@@ -206,16 +219,22 @@ static int fq_codel_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return NET_XMIT_SUCCESS;
 
 	prev_backlog = sch->qstats.backlog;
-	q->drop_overlimit++;
-	/* Return Congestion Notification only if we dropped a packet
-	 * from this flow.
-	 */
-	if (fq_codel_drop(sch) == idx)
-		return NET_XMIT_CN;
+	prev_qlen = sch->q.qlen;
 
-	/* As we dropped a packet, better let upper stack know this */
-	qdisc_tree_reduce_backlog(sch, 1, prev_backlog - sch->qstats.backlog);
-	return NET_XMIT_SUCCESS;
+	/* fq_codel_drop() is quite expensive, as it performs a linear search
+	 * in q->backlogs[] to find a fat flow.
+	 * So instead of dropping a single packet, drop half of its backlog
+	 * with a 64 packets limit to not add a too big cpu spike here.
+	 */
+	ret = fq_codel_drop(sch, q->drop_batch_size);
+
+	q->drop_overlimit += prev_qlen - sch->q.qlen;
+
+	/* As we dropped packet(s), better let upper stack know this */
+	qdisc_tree_reduce_backlog(sch, prev_qlen - sch->q.qlen,
+				  prev_backlog - sch->qstats.backlog);
+
+	return ret == idx ? NET_XMIT_CN : NET_XMIT_SUCCESS;
 }
 
 /* This is the specific function called from codel_dequeue()
@@ -335,6 +354,7 @@ static const struct nla_policy fq_codel_policy[TCA_FQ_CODEL_MAX + 1] = {
 	[TCA_FQ_CODEL_FLOWS]	= { .type = NLA_U32 },
 	[TCA_FQ_CODEL_QUANTUM]	= { .type = NLA_U32 },
 	[TCA_FQ_CODEL_CE_THRESHOLD] = { .type = NLA_U32 },
+	[TCA_FQ_CODEL_DROP_BATCH_SIZE] = { .type = NLA_U32 },
 };
 
 static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt)
@@ -386,6 +406,9 @@ static int fq_codel_change(struct Qdisc *sch, struct nlattr *opt)
 	if (tb[TCA_FQ_CODEL_QUANTUM])
 		q->quantum = max(256U, nla_get_u32(tb[TCA_FQ_CODEL_QUANTUM]));
 
+	if (tb[TCA_FQ_CODEL_DROP_BATCH_SIZE])
+		q->drop_batch_size = min(1U, nla_get_u32(tb[TCA_FQ_CODEL_DROP_BATCH_SIZE]));
+
 	while (sch->q.qlen > sch->limit) {
 		struct sk_buff *skb = fq_codel_dequeue(sch);
 
@@ -431,6 +454,7 @@ static int fq_codel_init(struct Qdisc *sch, struct nlattr *opt)
 
 	sch->limit = 10*1024;
 	q->flows_cnt = 1024;
+	q->drop_batch_size = 64;
 	q->quantum = psched_mtu(qdisc_dev(sch));
 	q->perturbation = prandom_u32();
 	INIT_LIST_HEAD(&q->new_flows);
@@ -489,6 +513,8 @@ static int fq_codel_dump(struct Qdisc *sch, struct sk_buff *skb)
 			q->cparams.ecn) ||
 	    nla_put_u32(skb, TCA_FQ_CODEL_QUANTUM,
 			q->quantum) ||
+	    nla_put_u32(skb, TCA_FQ_CODEL_DROP_BATCH_SIZE,
+			q->drop_batch_size) ||
 	    nla_put_u32(skb, TCA_FQ_CODEL_FLOWS,
 			q->flows_cnt))
 		goto nla_put_failure;