sch_htb: htb_requeue fix
Subsystems (3)
PATCH 1/1 sch_htb: htb_requeue fix
htb_requeue() enqueues skbs for which htb_classify() returns NULL. This is wrong because such skbs could be handled by NET_CLS_ACT code, and the decision could be different than earlier in htb_enqueue(). So htb_requeue() is changed to work and look more like htb_enqueue().
Signed-off-by Jarek Poplawski <[email protected]>
---
net/sched/sch_htb.c +10 -3
@@ -609,14 +609,14 @@ static int htb_enqueue(struct sk_buff *s/* TODO: requeuing packet charges it to policers again !! */ static int htb_requeue (struct sk_buff *skb,struct Qdisc *sch) { +int ret;struct htb_sched *q =qdisc_priv (sch); -int ret =NET_XMIT_SUCCESS ;struct htb_class *cl =htb_classify (skb, sch, &ret);struct sk_buff *tskb; -if (cl ==HTB_DIRECT || !cl) { +if (cl ==HTB_DIRECT ) {/* enqueue to helper queue */ -if (q->direct_queue.qlen < q->direct_qlen && cl) { +if (q->direct_queue.qlen < q->direct_qlen) {__skb_queue_head (&q->direct_queue, skb); }else {__skb_queue_head (&q->direct_queue, skb);
@@ -625,6 +625,13 @@ static int htb_requeue(struct sk_buff *s sch->qstats.drops++;return NET_XMIT_CN ; } +#ifdef CONFIG_NET_CLS_ACT + }else if (!cl) { +if (ret ==NET_XMIT_BYPASS ) + sch->qstats.drops++; +kfree_skb (skb); +return ret; +#endif }else if (cl->un.leaf.q->ops->requeue (skb, cl->un.leaf.q) !=NET_XMIT_SUCCESS ) { sch->qstats.drops++;
From: Jarek Poplawski <[email protected]>
Date: Fri, 8 Feb 2008 08:57:49 +0000
Quoted 7 lines
htb_requeue()enqueues skbs for whichhtb_classify()returns NULL. This is wrong because such skbs could be handled byNET_CLS_ACTcode, and the decision could be different than earlier inhtb_enqueue(). Sohtb_requeue()is changed to work and look more likehtb_enqueue().Signed-off-by Jarek Poplawski <[email protected]>
Applied, thanks.