[PATCH 3.16 193/305] net_sched: fix pfifo_head_drop behavior vs backlog

[Ben Hutchings]

3.16.37-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <edumazet@xxxxxxxxxx>

commit 6c0d54f1897d229748d4f41ef919078db6db2123 upstream.

When the qdisc is full, we drop a packet at the head of the queue,
queue the current skb and return NET_XMIT_CN

Now we track backlog on upper qdiscs, we need to call
qdisc_tree_reduce_backlog(), even if the qlen did not change.

Fixes: 2ccccf5fb43f ("net_sched: update hierarchical backlog too")
Signed-off-by: Eric Dumazet <edumazet@xxxxxxxxxx>
Cc: WANG Cong <xiyou.wangcong@xxxxxxxxx>
Cc: Jamal Hadi Salim <jhs@xxxxxxxxxxxx>
Acked-by: Cong Wang <xiyou.wangcong@xxxxxxxxx>
Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx>
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
 net/sched/sch_fifo.c | 4 ++++
 1 file changed, 4 insertions(+)

--- a/net/sched/sch_fifo.c
+++ b/net/sched/sch_fifo.c
@@ -37,14 +37,18 @@ static int pfifo_enqueue(struct sk_buff
 
 static int pfifo_tail_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 {
+	unsigned int prev_backlog;
+
 	if (likely(skb_queue_len(&sch->q) < sch->limit))
 		return qdisc_enqueue_tail(skb, sch);
 
+	prev_backlog = sch->qstats.backlog;
 	/* queue full, remove one skb to fulfill the limit */
 	__qdisc_queue_drop_head(sch, &sch->q);
 	sch->qstats.drops++;
 	qdisc_enqueue_tail(skb, sch);
 
+	qdisc_tree_reduce_backlog(sch, 0, prev_backlog - sch->qstats.backlog);
 	return NET_XMIT_CN;
 }