net_sched: dsmark: use qdisc_dequeue_peeked()
authorKyeong Yoo <Kyeong.Yoo@alliedtelesis.co.nz>
Mon, 7 Mar 2016 04:07:57 +0000 (17:07 +1300)
committerDavid S. Miller <davem@davemloft.net>
Tue, 8 Mar 2016 19:35:13 +0000 (14:35 -0500)
commitf8b33d8e870758ccff13e5f81fd5050b52a42d35
tree4d38ca0aff08cc8aa392761d465c3fa232e65ff1
parent4c38cd61aef20fce34c669caa901634ca5f88bf8
net_sched: dsmark: use qdisc_dequeue_peeked()

This fix is for dsmark similar to commit 3557619f0f6f7496ed453d4825e249
("net_sched: prio: use qdisc_dequeue_peeked")
and makes use of qdisc_dequeue_peeked() instead of direct dequeue() call.

First time, wrr peeks dsmark, which will then peek into sfq.
sfq dequeues an skb and it's stored in sch->gso_skb.
Next time, wrr tries to dequeue from dsmark, which will call sfq dequeue
directly. This results skipping the previously peeked skb.

So changed dsmark dequeue to call qdisc_dequeue_peeked() instead to use
peeked skb if exists.

Signed-off-by: Kyeong Yoo <kyeong.yoo@alliedtelesis.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/sched/sch_dsmark.c