Merge branch 'linux-2.6' into for-linus

25 files changed, 189 insertions, 111 deletions

diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index 8298ea9ffe1..f4544dd8647 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -6,6 +6,7 @@ menu "QoS and/or fair queueing"
 
 config NET_SCHED
 	bool "QoS and/or fair queueing"
+	select NET_SCH_FIFO
 	---help---
 	  When the kernel has several packets to send out over a network
 	  device, it has to decide which ones to send first, which ones to
@@ -40,6 +41,9 @@ config NET_SCHED
 	  The available schedulers are listed in the following questions; you
 	  can say Y to as many as you like. If unsure, say N now.
 
+config NET_SCH_FIFO
+	bool
+
 if NET_SCHED
 
 choice
@@ -320,7 +324,7 @@ config CLS_U32_PERF
 
 config CLS_U32_MARK
 	bool "Netfilter marks support"
-	depends on NET_CLS_U32 && NETFILTER
+	depends on NET_CLS_U32
 	---help---
 	  Say Y here to be able to use netfilter marks as u32 key.
 
diff --git a/net/sched/Makefile b/net/sched/Makefile
index 0f06aec6609..ff2d6e5e282 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -4,7 +4,7 @@
 
 obj-y	:= sch_generic.o
 
-obj-$(CONFIG_NET_SCHED)		+= sch_api.o sch_fifo.o sch_blackhole.o
+obj-$(CONFIG_NET_SCHED)		+= sch_api.o sch_blackhole.o
 obj-$(CONFIG_NET_CLS)		+= cls_api.o
 obj-$(CONFIG_NET_CLS_ACT)	+= act_api.o
 obj-$(CONFIG_NET_ACT_POLICE)	+= act_police.o
@@ -14,6 +14,7 @@ obj-$(CONFIG_NET_ACT_MIRRED)	+= act_mirred.o
 obj-$(CONFIG_NET_ACT_IPT)	+= act_ipt.o
 obj-$(CONFIG_NET_ACT_PEDIT)	+= act_pedit.o
 obj-$(CONFIG_NET_ACT_SIMP)	+= act_simple.o
+obj-$(CONFIG_NET_SCH_FIFO)	+= sch_fifo.o
 obj-$(CONFIG_NET_SCH_CBQ)	+= sch_cbq.o
 obj-$(CONFIG_NET_SCH_HTB)	+= sch_htb.o
 obj-$(CONFIG_NET_SCH_HPFQ)	+= sch_hpfq.o
diff --git a/net/sched/act_gact.c b/net/sched/act_gact.c
index 6cff56696a8..85de7efd5fe 100644
--- a/net/sched/act_gact.c
+++ b/net/sched/act_gact.c
@@ -48,14 +48,14 @@ static struct tcf_hashinfo gact_hash_info = {
 #ifdef CONFIG_GACT_PROB
 static int gact_net_rand(struct tcf_gact *gact)
 {
-	if (net_random() % gact->tcfg_pval)
+	if (!gact->tcfg_pval || net_random() % gact->tcfg_pval)
 		return gact->tcf_action;
 	return gact->tcfg_paction;
 }
 
 static int gact_determ(struct tcf_gact *gact)
 {
-	if (gact->tcf_bstats.packets % gact->tcfg_pval)
+	if (!gact->tcfg_pval || gact->tcf_bstats.packets % gact->tcfg_pval)
 		return gact->tcf_action;
 	return gact->tcfg_paction;
 }
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index d8c9310da6e..a9608064a4c 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -156,10 +156,9 @@ static int tcf_ipt_init(struct rtattr *rta, struct rtattr *est,
 	    rtattr_strlcpy(tname, tb[TCA_IPT_TABLE-1], IFNAMSIZ) >= IFNAMSIZ)
 		strcpy(tname, "mangle");
 
-	t = kmalloc(td->u.target_size, GFP_KERNEL);
+	t = kmemdup(td, td->u.target_size, GFP_KERNEL);
 	if (unlikely(!t))
 		goto err2;
-	memcpy(t, td, td->u.target_size);
 
 	if ((err = ipt_init_target(t, tname, hook)) < 0)
 		goto err3;
@@ -256,13 +255,12 @@ static int tcf_ipt_dump(struct sk_buff *skb, struct tc_action *a, int bind, int
 	** for foolproof you need to not assume this
 	*/
 
-	t = kmalloc(ipt->tcfi_t->u.user.target_size, GFP_ATOMIC);
+	t = kmemdup(ipt->tcfi_t, ipt->tcfi_t->u.user.target_size, GFP_ATOMIC);
 	if (unlikely(!t))
 		goto rtattr_failure;
 
 	c.bindcnt = ipt->tcf_bindcnt - bind;
 	c.refcnt = ipt->tcf_refcnt - ref;
-	memcpy(t, ipt->tcfi_t, ipt->tcfi_t->u.user.target_size);
 	strcpy(t->u.user.name, ipt->tcfi_t->u.kernel.target->name);
 
 	RTA_PUT(skb, TCA_IPT_TARG, ipt->tcfi_t->u.user.target_size, t);
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index fed47b65883..af68e1e8325 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -46,6 +46,18 @@ static struct tcf_hashinfo police_hash_info = {
 	.lock	=	&police_lock,
 };
 
+/* old policer structure from before tc actions */
+struct tc_police_compat
+{
+	u32			index;
+	int			action;
+	u32			limit;
+	u32			burst;
+	u32			mtu;
+	struct tc_ratespec	rate;
+	struct tc_ratespec	peakrate;
+};
+
 /* Each policer is serialized by its individual spinlock */
 
 #ifdef CONFIG_NET_CLS_ACT
@@ -131,12 +143,15 @@ static int tcf_act_police_locate(struct rtattr *rta, struct rtattr *est,
 	struct tc_police *parm;
 	struct tcf_police *police;
 	struct qdisc_rate_table *R_tab = NULL, *P_tab = NULL;
+	int size;
 
 	if (rta == NULL || rtattr_parse_nested(tb, TCA_POLICE_MAX, rta) < 0)
 		return -EINVAL;
 
-	if (tb[TCA_POLICE_TBF-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_POLICE_TBF-1]) != sizeof(*parm))
+	if (tb[TCA_POLICE_TBF-1] == NULL)
+		return -EINVAL;
+	size = RTA_PAYLOAD(tb[TCA_POLICE_TBF-1]);
+	if (size != sizeof(*parm) && size != sizeof(struct tc_police_compat))
 		return -EINVAL;
 	parm = RTA_DATA(tb[TCA_POLICE_TBF-1]);
 
@@ -415,12 +430,15 @@ struct tcf_police *tcf_police_locate(struct rtattr *rta, struct rtattr *est)
 	struct tcf_police *police;
 	struct rtattr *tb[TCA_POLICE_MAX];
 	struct tc_police *parm;
+	int size;
 
 	if (rtattr_parse_nested(tb, TCA_POLICE_MAX, rta) < 0)
 		return NULL;
 
-	if (tb[TCA_POLICE_TBF-1] == NULL ||
-	    RTA_PAYLOAD(tb[TCA_POLICE_TBF-1]) != sizeof(*parm))
+	if (tb[TCA_POLICE_TBF-1] == NULL)
+		return NULL;
+	size = RTA_PAYLOAD(tb[TCA_POLICE_TBF-1]);
+	if (size != sizeof(*parm) && size != sizeof(struct tc_police_compat))
 		return NULL;
 
 	parm = RTA_DATA(tb[TCA_POLICE_TBF-1]);
diff --git a/net/sched/act_simple.c b/net/sched/act_simple.c
index 901571a6770..5fe80854ca9 100644
--- a/net/sched/act_simple.c
+++ b/net/sched/act_simple.c
@@ -71,11 +71,10 @@ static int tcf_simp_release(struct tcf_defact *d, int bind)
 
 static int alloc_defdata(struct tcf_defact *d, u32 datalen, void *defdata)
 {
-	d->tcfd_defdata = kmalloc(datalen, GFP_KERNEL);
+	d->tcfd_defdata = kmemdup(defdata, datalen, GFP_KERNEL);
 	if (unlikely(!d->tcfd_defdata))
 		return -ENOMEM;
 	d->tcfd_datalen = datalen;
-	memcpy(d->tcfd_defdata, defdata, datalen);
 	return 0;
 }
 
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index 37a18402164..edb8fc97ae1 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -217,7 +217,7 @@ replay:
 		/* Create new proto tcf */
 
 		err = -ENOBUFS;
-		if ((tp = kmalloc(sizeof(*tp), GFP_KERNEL)) == NULL)
+		if ((tp = kzalloc(sizeof(*tp), GFP_KERNEL)) == NULL)
 			goto errout;
 		err = -EINVAL;
 		tp_ops = tcf_proto_lookup_ops(tca[TCA_KIND-1]);
@@ -247,7 +247,6 @@ replay:
 			kfree(tp);
 			goto errout;
 		}
-		memset(tp, 0, sizeof(*tp));
 		tp->ops = tp_ops;
 		tp->protocol = protocol;
 		tp->prio = nprio ? : tcf_auto_prio(*back);
diff --git a/net/sched/cls_fw.c b/net/sched/cls_fw.c
index e54acc6bccc..f59a2c4aa03 100644
--- a/net/sched/cls_fw.c
+++ b/net/sched/cls_fw.c
@@ -101,11 +101,7 @@ static int fw_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	struct fw_head *head = (struct fw_head*)tp->root;
 	struct fw_filter *f;
 	int r;
-#ifdef CONFIG_NETFILTER
-	u32 id = skb->nfmark & head->mask;
-#else
-	u32 id = 0;
-#endif
+	u32 id = skb->mark & head->mask;
 
 	if (head != NULL) {
 		for (f=head->ht[fw_hash(id)]; f; f=f->next) {
diff --git a/net/sched/cls_rsvp.h b/net/sched/cls_rsvp.h
index 6e230ecfba0..587b9adab38 100644
--- a/net/sched/cls_rsvp.h
+++ b/net/sched/cls_rsvp.h
@@ -77,7 +77,7 @@ struct rsvp_head
 struct rsvp_session
 {
 	struct rsvp_session	*next;
-	u32			dst[RSVP_DST_LEN];
+	__be32			dst[RSVP_DST_LEN];
 	struct tc_rsvp_gpi 	dpi;
 	u8			protocol;
 	u8			tunnelid;
@@ -89,7 +89,7 @@ struct rsvp_session
 struct rsvp_filter
 {
 	struct rsvp_filter	*next;
-	u32			src[RSVP_DST_LEN];
+	__be32			src[RSVP_DST_LEN];
 	struct tc_rsvp_gpi	spi;
 	u8			tunnelhdr;
 
@@ -100,17 +100,17 @@ struct rsvp_filter
 	struct rsvp_session	*sess;
 };
 
-static __inline__ unsigned hash_dst(u32 *dst, u8 protocol, u8 tunnelid)
+static __inline__ unsigned hash_dst(__be32 *dst, u8 protocol, u8 tunnelid)
 {
-	unsigned h = dst[RSVP_DST_LEN-1];
+	unsigned h = (__force __u32)dst[RSVP_DST_LEN-1];
 	h ^= h>>16;
 	h ^= h>>8;
 	return (h ^ protocol ^ tunnelid) & 0xFF;
 }
 
-static __inline__ unsigned hash_src(u32 *src)
+static __inline__ unsigned hash_src(__be32 *src)
 {
-	unsigned h = src[RSVP_DST_LEN-1];
+	unsigned h = (__force __u32)src[RSVP_DST_LEN-1];
 	h ^= h>>16;
 	h ^= h>>8;
 	h ^= h>>4;
@@ -138,7 +138,7 @@ static int rsvp_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	struct rsvp_session *s;
 	struct rsvp_filter *f;
 	unsigned h1, h2;
-	u32 *dst, *src;
+	__be32 *dst, *src;
 	u8 protocol;
 	u8 tunnelid = 0;
 	u8 *xprt;
@@ -410,7 +410,7 @@ static int rsvp_change(struct tcf_proto *tp, unsigned long base,
 	struct rtattr *tb[TCA_RSVP_MAX];
 	struct tcf_exts e;
 	unsigned h1, h2;
-	u32 *dst;
+	__be32 *dst;
 	int err;
 
 	if (opt == NULL)
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 0a6cfa0005b..8b519480199 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -143,7 +143,7 @@ next_knode:
 #endif
 
 #ifdef CONFIG_CLS_U32_MARK
-		if ((skb->nfmark & n->mark.mask) != n->mark.val) {
+		if ((skb->mark & n->mark.mask) != n->mark.val) {
 			n = n->next;
 			goto next_knode;
 		} else {
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index 61e3b740ab1..45d47d37155 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -208,13 +208,9 @@ META_COLLECTOR(int_maclen)
  * Netfilter
  **************************************************************************/
 
-META_COLLECTOR(int_nfmark)
+META_COLLECTOR(int_mark)
 {
-#ifdef CONFIG_NETFILTER
-	dst->value = skb->nfmark;
-#else
-	dst->value = 0;
-#endif
+	dst->value = skb->mark;
 }
 
 /**************************************************************************
@@ -490,7 +486,7 @@ static struct meta_ops __meta_ops[TCF_META_TYPE_MAX+1][TCF_META_ID_MAX+1] = {
 		[META_ID(PKTLEN)]		= META_FUNC(int_pktlen),
 		[META_ID(DATALEN)]		= META_FUNC(int_datalen),
 		[META_ID(MACLEN)]		= META_FUNC(int_maclen),
-		[META_ID(NFMARK)]		= META_FUNC(int_nfmark),
+		[META_ID(NFMARK)]		= META_FUNC(int_mark),
 		[META_ID(TCINDEX)]		= META_FUNC(int_tcindex),
 		[META_ID(RTCLASSID)]		= META_FUNC(int_rtclassid),
 		[META_ID(RTIIF)]		= META_FUNC(int_rtiif),
@@ -550,10 +546,9 @@ static int meta_var_change(struct meta_value *dst, struct rtattr *rta)
 {
 	int len = RTA_PAYLOAD(rta);
 
-	dst->val = (unsigned long) kmalloc(len, GFP_KERNEL);
+	dst->val = (unsigned long)kmemdup(RTA_DATA(rta), len, GFP_KERNEL);
 	if (dst->val == 0UL)
 		return -ENOMEM;
-	memcpy((void *) dst->val, RTA_DATA(rta), len);
 	dst->len = len;
 	return 0;
 }
diff --git a/net/sched/em_nbyte.c b/net/sched/em_nbyte.c
index cc80babfd79..005db409be6 100644
--- a/net/sched/em_nbyte.c
+++ b/net/sched/em_nbyte.c
@@ -34,12 +34,10 @@ static int em_nbyte_change(struct tcf_proto *tp, void *data, int data_len,
 		return -EINVAL;
 
 	em->datalen = sizeof(*nbyte) + nbyte->len;
-	em->data = (unsigned long) kmalloc(em->datalen, GFP_KERNEL);
+	em->data = (unsigned long)kmemdup(data, em->datalen, GFP_KERNEL);
 	if (em->data == 0UL)
 		return -ENOBUFS;
 
-	memcpy((void *) em->data, data, em->datalen);
-
 	return 0;
 }
 
diff --git a/net/sched/ematch.c b/net/sched/ematch.c
index 0fd0768a17c..8f8a16da72a 100644
--- a/net/sched/ematch.c
+++ b/net/sched/ematch.c
@@ -251,12 +251,11 @@ static int tcf_em_validate(struct tcf_proto *tp,
 					goto errout;
 				em->data = *(u32 *) data;
 			} else {
-				void *v = kmalloc(data_len, GFP_KERNEL);
+				void *v = kmemdup(data, data_len, GFP_KERNEL);
 				if (v == NULL) {
 					err = -ENOBUFS;
 					goto errout;
 				}
-				memcpy(v, data, data_len);
 				em->data = (unsigned long) v;
 			}
 		}
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 0b648929114..65825f4409d 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -191,21 +191,27 @@ int unregister_qdisc(struct Qdisc_ops *qops)
    (root qdisc, all its children, children of children etc.)
  */
 
-struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+static struct Qdisc *__qdisc_lookup(struct net_device *dev, u32 handle)
 {
 	struct Qdisc *q;
 
-	read_lock(&qdisc_tree_lock);
 	list_for_each_entry(q, &dev->qdisc_list, list) {
-		if (q->handle == handle) {
-			read_unlock(&qdisc_tree_lock);
+		if (q->handle == handle)
 			return q;
-		}
 	}
-	read_unlock(&qdisc_tree_lock);
 	return NULL;
 }
 
+struct Qdisc *qdisc_lookup(struct net_device *dev, u32 handle)
+{
+	struct Qdisc *q;
+
+	read_lock(&qdisc_tree_lock);
+	q = __qdisc_lookup(dev, handle);
+	read_unlock(&qdisc_tree_lock);
+	return q;
+}
+
 static struct Qdisc *qdisc_leaf(struct Qdisc *p, u32 classid)
 {
 	unsigned long cl;
@@ -348,6 +354,26 @@ dev_graft_qdisc(struct net_device *dev, struct Qdisc *qdisc)
 	return oqdisc;
 }
 
+void qdisc_tree_decrease_qlen(struct Qdisc *sch, unsigned int n)
+{
+	struct Qdisc_class_ops *cops;
+	unsigned long cl;
+	u32 parentid;
+
+	if (n == 0)
+		return;
+	while ((parentid = sch->parent)) {
+		sch = __qdisc_lookup(sch->dev, TC_H_MAJ(parentid));
+		cops = sch->ops->cl_ops;
+		if (cops->qlen_notify) {
+			cl = cops->get(sch, parentid);
+			cops->qlen_notify(sch, cl);
+			cops->put(sch, cl);
+		}
+		sch->q.qlen -= n;
+	}
+}
+EXPORT_SYMBOL(qdisc_tree_decrease_qlen);
 
 /* Graft qdisc "new" to class "classid" of qdisc "parent" or
    to device "dev".
@@ -1112,7 +1138,7 @@ int tc_classify(struct sk_buff *skb, struct tcf_proto *tp,
 	struct tcf_result *res)
 {
 	int err = 0;
-	u32 protocol = skb->protocol;
+	__be16 protocol = skb->protocol;
 #ifdef CONFIG_NET_CLS_ACT
 	struct tcf_proto *otp = tp;
 reclassify:
@@ -1277,7 +1303,6 @@ static int __init pktsched_init(void)
 
 subsys_initcall(pktsched_init);
 
-EXPORT_SYMBOL(qdisc_lookup);
 EXPORT_SYMBOL(qdisc_get_rtab);
 EXPORT_SYMBOL(qdisc_put_rtab);
 EXPORT_SYMBOL(register_qdisc);
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index dbf44da0912..edc7bb0b9c8 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -316,7 +316,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
 	}
 	memset(flow,0,sizeof(*flow));
 	flow->filter_list = NULL;
-	if (!(flow->q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops)))
+	if (!(flow->q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops,classid)))
 		flow->q = &noop_qdisc;
 	DPRINTK("atm_tc_change: qdisc %p\n",flow->q);
 	flow->sock = sock;
@@ -576,7 +576,8 @@ static int atm_tc_init(struct Qdisc *sch,struct rtattr *opt)
 
 	DPRINTK("atm_tc_init(sch %p,[qdisc %p],opt %p)\n",sch,p,opt);
 	p->flows = &p->link;
-	if(!(p->link.q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops)))
+	if(!(p->link.q = qdisc_create_dflt(sch->dev,&pfifo_qdisc_ops,
+					   sch->handle)))
 		p->link.q = &noop_qdisc;
 	DPRINTK("atm_tc_init: link (%p) qdisc %p\n",&p->link,p->link.q);
 	p->link.filter_list = NULL;
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index bac881bfe36..ba82dfab604 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1429,7 +1429,8 @@ static int cbq_init(struct Qdisc *sch, struct rtattr *opt)
 	q->link.sibling = &q->link;
 	q->link.classid = sch->handle;
 	q->link.qdisc = sch;
-	if (!(q->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops)))
+	if (!(q->link.q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+					    sch->handle)))
 		q->link.q = &noop_qdisc;
 
 	q->link.priority = TC_CBQ_MAXPRIO-1;
@@ -1674,7 +1675,8 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	if (cl) {
 		if (new == NULL) {
-			if ((new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops)) == NULL)
+			if ((new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+						     cl->classid)) == NULL)
 				return -ENOBUFS;
 		} else {
 #ifdef CONFIG_NET_CLS_POLICE
@@ -1685,7 +1687,7 @@ static int cbq_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 		sch_tree_lock(sch);
 		*old = cl->q;
 		cl->q = new;
-		sch->q.qlen -= (*old)->q.qlen;
+		qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 		qdisc_reset(*old);
 		sch_tree_unlock(sch);
 
@@ -1932,7 +1934,7 @@ cbq_change_class(struct Qdisc *sch, u32 classid, u32 parentid, struct rtattr **t
 	cl->R_tab = rtab;
 	rtab = NULL;
 	cl->refcnt = 1;
-	if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops)))
+	if (!(cl->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid)))
 		cl->q = &noop_qdisc;
 	cl->classid = classid;
 	cl->tparent = parent;
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index 11c8a2119b9..d5421816f00 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -88,15 +88,16 @@ static int dsmark_graft(struct Qdisc *sch, unsigned long arg,
 		sch, p, new, old);
 
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+					sch->handle);
 		if (new == NULL)
 			new = &noop_qdisc;
 	}
 
 	sch_tree_lock(sch);
 	*old = xchg(&p->q, new);
+	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 	qdisc_reset(*old);
-	sch->q.qlen = 0;
 	sch_tree_unlock(sch);
 
         return 0;
@@ -307,7 +308,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
 			if (p->mask[index] != 0xff || p->value[index])
 				printk(KERN_WARNING "dsmark_dequeue: "
 				       "unsupported protocol %d\n",
-				       htons(skb->protocol));
+				       ntohs(skb->protocol));
 			break;
 	};
 
@@ -387,7 +388,7 @@ static int dsmark_init(struct Qdisc *sch, struct rtattr *opt)
 	p->default_index = default_index;
 	p->set_tc_index = RTA_GET_FLAG(tb[TCA_DSMARK_SET_TC_INDEX-1]);
 
-	p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+	p->q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, sch->handle);
 	if (p->q == NULL)
 		p->q = &noop_qdisc;
 
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 88c6a99ce53..bc116bd6937 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -450,13 +450,15 @@ errout:
 	return ERR_PTR(-err);
 }
 
-struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops)
+struct Qdisc * qdisc_create_dflt(struct net_device *dev, struct Qdisc_ops *ops,
+				 unsigned int parentid)
 {
 	struct Qdisc *sch;
 	
 	sch = qdisc_alloc(dev, ops);
 	if (IS_ERR(sch))
 		goto errout;
+	sch->parent = parentid;
 
 	if (!ops->init || ops->init(sch, NULL) == 0)
 		return sch;
@@ -520,7 +522,8 @@ void dev_activate(struct net_device *dev)
 	if (dev->qdisc_sleeping == &noop_qdisc) {
 		struct Qdisc *qdisc;
 		if (dev->tx_queue_len) {
-			qdisc = qdisc_create_dflt(dev, &pfifo_fast_ops);
+			qdisc = qdisc_create_dflt(dev, &pfifo_fast_ops,
+						  TC_H_ROOT);
 			if (qdisc == NULL) {
 				printk(KERN_INFO "%s: activation failed\n", dev->name);
 				return;
@@ -606,13 +609,10 @@ void dev_shutdown(struct net_device *dev)
 	qdisc_unlock_tree(dev);
 }
 
-EXPORT_SYMBOL(__netdev_watchdog_up);
 EXPORT_SYMBOL(netif_carrier_on);
 EXPORT_SYMBOL(netif_carrier_off);
 EXPORT_SYMBOL(noop_qdisc);
-EXPORT_SYMBOL(noop_qdisc_ops);
 EXPORT_SYMBOL(qdisc_create_dflt);
-EXPORT_SYMBOL(qdisc_alloc);
 EXPORT_SYMBOL(qdisc_destroy);
 EXPORT_SYMBOL(qdisc_reset);
 EXPORT_SYMBOL(qdisc_lock_tree);
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index 6a6735a2ed3..6eefa699577 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -946,6 +946,7 @@ qdisc_peek_len(struct Qdisc *sch)
 	if (unlikely(sch->ops->requeue(skb, sch) != NET_XMIT_SUCCESS)) {
 		if (net_ratelimit())
 			printk("qdisc_peek_len: failed to requeue\n");
+		qdisc_tree_decrease_qlen(sch, 1);
 		return 0;
 	}
 	return len;
@@ -957,11 +958,7 @@ hfsc_purge_queue(struct Qdisc *sch, struct hfsc_class *cl)
 	unsigned int len = cl->qdisc->q.qlen;
 
 	qdisc_reset(cl->qdisc);
-	if (len > 0) {
-		update_vf(cl, 0, 0);
-		set_passive(cl);
-		sch->q.qlen -= len;
-	}
+	qdisc_tree_decrease_qlen(cl->qdisc, len);
 }
 
 static void
@@ -1138,7 +1135,7 @@ hfsc_change_class(struct Qdisc *sch, u32 classid, u32 parentid,
 	cl->classid   = classid;
 	cl->sched     = q;
 	cl->cl_parent = parent;
-	cl->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+	cl->qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
 	if (cl->qdisc == NULL)
 		cl->qdisc = &noop_qdisc;
 	cl->stats_lock = &sch->dev->queue_lock;
@@ -1271,7 +1268,8 @@ hfsc_graft_class(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	if (cl->level > 0)
 		return -EINVAL;
 	if (new == NULL) {
-		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+		new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+					cl->classid);
 		if (new == NULL)
 			new = &noop_qdisc;
 	}
@@ -1294,6 +1292,17 @@ hfsc_class_leaf(struct Qdisc *sch, unsigned long arg)
 	return NULL;
 }
 
+static void
+hfsc_qlen_notify(struct Qdisc *sch, unsigned long arg)
+{
+	struct hfsc_class *cl = (struct hfsc_class *)arg;
+
+	if (cl->qdisc->q.qlen == 0) {
+		update_vf(cl, 0, 0);
+		set_passive(cl);
+	}
+}
+
 static unsigned long
 hfsc_get_class(struct Qdisc *sch, u32 classid)
 {
@@ -1514,7 +1523,8 @@ hfsc_init_qdisc(struct Qdisc *sch, struct rtattr *opt)
 	q->root.refcnt  = 1;
 	q->root.classid = sch->handle;
 	q->root.sched   = q;
-	q->root.qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+	q->root.qdisc = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+					  sch->handle);
 	if (q->root.qdisc == NULL)
 		q->root.qdisc = &noop_qdisc;
 	q->root.stats_lock = &sch->dev->queue_lock;
@@ -1777,6 +1787,7 @@ static struct Qdisc_class_ops hfsc_class_ops = {
 	.delete		= hfsc_delete_class,
 	.graft		= hfsc_graft_class,
 	.leaf		= hfsc_class_leaf,
+	.qlen_notify	= hfsc_qlen_notify,
 	.get		= hfsc_get_class,
 	.put		= hfsc_put_class,
 	.bind_tcf	= hfsc_bind_tcf,
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 9b9c555c713..215e68c2b61 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -1223,17 +1223,14 @@ static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	struct htb_class *cl = (struct htb_class *)arg;
 
 	if (cl && !cl->level) {
-		if (new == NULL && (new = qdisc_create_dflt(sch->dev,
-							    &pfifo_qdisc_ops))
+		if (new == NULL &&
+		    (new = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+		    			     cl->classid))
 		    == NULL)
 			return -ENOBUFS;
 		sch_tree_lock(sch);
 		if ((*old = xchg(&cl->un.leaf.q, new)) != NULL) {
-			if (cl->prio_activity)
-				htb_deactivate(qdisc_priv(sch), cl);
-
-			/* TODO: is it correct ? Why CBQ doesn't do it ? */
-			sch->q.qlen -= (*old)->q.qlen;
+			qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 			qdisc_reset(*old);
 		}
 		sch_tree_unlock(sch);
@@ -1248,6 +1245,14 @@ static struct Qdisc *htb_leaf(struct Qdisc *sch, unsigned long arg)
 	return (cl && !cl->level) ? cl->un.leaf.q : NULL;
 }
 
+static void htb_qlen_notify(struct Qdisc *sch, unsigned long arg)
+{
+	struct htb_class *cl = (struct htb_class *)arg;
+
+	if (cl->un.leaf.q->q.qlen == 0)
+		htb_deactivate(qdisc_priv(sch), cl);
+}
+
 static unsigned long htb_get(struct Qdisc *sch, u32 classid)
 {
 	struct htb_class *cl = htb_find(classid, sch);
@@ -1269,9 +1274,9 @@ static void htb_destroy_filters(struct tcf_proto **fl)
 static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl)
 {
 	struct htb_sched *q = qdisc_priv(sch);
+
 	if (!cl->level) {
 		BUG_TRAP(cl->un.leaf.q);
-		sch->q.qlen -= cl->un.leaf.q->q.qlen;
 		qdisc_destroy(cl->un.leaf.q);
 	}
 	qdisc_put_rtab(cl->rate);
@@ -1284,8 +1289,7 @@ static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl)
 						  struct htb_class, sibling));
 
 	/* note: this delete may happen twice (see htb_delete) */
-	if (!hlist_unhashed(&cl->hlist))
-		hlist_del(&cl->hlist);
+	hlist_del_init(&cl->hlist);
 	list_del(&cl->sibling);
 
 	if (cl->prio_activity)
@@ -1323,6 +1327,7 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 {
 	struct htb_sched *q = qdisc_priv(sch);
 	struct htb_class *cl = (struct htb_class *)arg;
+	unsigned int qlen;
 
 	// TODO: why don't allow to delete subtree ? references ? does
 	// tc subsys quarantee us that in htb_destroy it holds no class
@@ -1333,8 +1338,13 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg)
 	sch_tree_lock(sch);
 
 	/* delete from hash and active; remainder in destroy_class */
-	if (!hlist_unhashed(&cl->hlist))
-		hlist_del(&cl->hlist);
+	hlist_del_init(&cl->hlist);
+
+	if (!cl->level) {
+		qlen = cl->un.leaf.q->q.qlen;
+		qdisc_reset(cl->un.leaf.q);
+		qdisc_tree_decrease_qlen(cl->un.leaf.q, qlen);
+	}
 
 	if (cl->prio_activity)
 		htb_deactivate(q, cl);
@@ -1412,11 +1422,14 @@ static int htb_change_class(struct Qdisc *sch, u32 classid,
 		/* create leaf qdisc early because it uses kmalloc(GFP_KERNEL)
 		   so that can't be used inside of sch_tree_lock
 		   -- thanks to Karlis Peisenieks */
-		new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+		new_q = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops, classid);
 		sch_tree_lock(sch);
 		if (parent && !parent->level) {
+			unsigned int qlen = parent->un.leaf.q->q.qlen;
+
 			/* turn parent into inner node */
-			sch->q.qlen -= parent->un.leaf.q->q.qlen;
+			qdisc_reset(parent->un.leaf.q);
+			qdisc_tree_decrease_qlen(parent->un.leaf.q, qlen);
 			qdisc_destroy(parent->un.leaf.q);
 			if (parent->prio_activity)
 				htb_deactivate(q, parent);
@@ -1564,6 +1577,7 @@ static void htb_walk(struct Qdisc *sch, struct qdisc_walker *arg)
 static struct Qdisc_class_ops htb_class_ops = {
 	.graft		=	htb_graft,
 	.leaf		=	htb_leaf,
+	.qlen_notify	=	htb_qlen_notify,
 	.get		=	htb_get,
 	.put		=	htb_put,
 	.change		=	htb_change_class,
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 45939bafbdf..79542af9dab 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -4,7 +4,7 @@
  * 		This program is free software; you can redistribute it and/or
  * 		modify it under the terms of the GNU General Public License
  * 		as published by the Free Software Foundation; either version
- * 		2 of the License, or (at your option) any later version.
+ * 		2 of the License.
  *
  *  		Many of the algorithms and ideas for this came from
  *		NIST Net which is not copyrighted. 
@@ -170,6 +170,8 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch)
 		return NET_XMIT_BYPASS;
 	}
 
+	skb_orphan(skb);
+
 	/*
 	 * If we need to duplicate packet, then re-insert at top of the
 	 * qdisc tree, since parent queuer expects that only one
@@ -285,13 +287,10 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch)
 			psched_tdiff_t delay = PSCHED_TDIFF(cb->time_to_send, now);
 
 			if (q->qdisc->ops->requeue(skb, q->qdisc) != NET_XMIT_SUCCESS) {
+				qdisc_tree_decrease_qlen(q->qdisc, 1);
 				sch->qstats.drops++;
-
-				/* After this qlen is confused */
 				printk(KERN_ERR "netem: queue discpline %s could not requeue\n",
 				       q->qdisc->ops->id);
-
-				sch->q.qlen--;
 			}
 
 			mod_timer(&q->timer, jiffies + PSCHED_US2JIFFIE(delay));
@@ -572,7 +571,8 @@ static int netem_init(struct Qdisc *sch, struct rtattr *opt)
 	q->timer.function = netem_watchdog;
 	q->timer.data = (unsigned long) sch;
 
-	q->qdisc = qdisc_create_dflt(sch->dev, &tfifo_qdisc_ops);
+	q->qdisc = qdisc_create_dflt(sch->dev, &tfifo_qdisc_ops,
+				     TC_H_MAKE(sch->handle, 1));
 	if (!q->qdisc) {
 		pr_debug("netem: qdisc create failed\n");
 		return -ENOMEM;
@@ -659,8 +659,8 @@ static int netem_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	sch_tree_lock(sch);
 	*old = xchg(&q->qdisc, new);
+	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 	qdisc_reset(*old);
-	sch->q.qlen = 0;
 	sch_tree_unlock(sch);
 
 	return 0;
diff --git a/net/sched/sch_prio.c b/net/sched/sch_prio.c
index a5fa03c0c19..2567b4c96c1 100644
--- a/net/sched/sch_prio.c
+++ b/net/sched/sch_prio.c
@@ -222,21 +222,27 @@ static int prio_tune(struct Qdisc *sch, struct rtattr *opt)
 
 	for (i=q->bands; i<TCQ_PRIO_BANDS; i++) {
 		struct Qdisc *child = xchg(&q->queues[i], &noop_qdisc);
-		if (child != &noop_qdisc)
+		if (child != &noop_qdisc) {
+			qdisc_tree_decrease_qlen(child, child->q.qlen);
 			qdisc_destroy(child);
+		}
 	}
 	sch_tree_unlock(sch);
 
 	for (i=0; i<q->bands; i++) {
 		if (q->queues[i] == &noop_qdisc) {
 			struct Qdisc *child;
-			child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops);
+			child = qdisc_create_dflt(sch->dev, &pfifo_qdisc_ops,
+						  TC_H_MAKE(sch->handle, i + 1));
 			if (child) {
 				sch_tree_lock(sch);
 				child = xchg(&q->queues[i], child);
 
-				if (child != &noop_qdisc)
+				if (child != &noop_qdisc) {
+					qdisc_tree_decrease_qlen(child,
+								 child->q.qlen);
 					qdisc_destroy(child);
+				}
 				sch_tree_unlock(sch);
 			}
 		}
@@ -294,7 +300,7 @@ static int prio_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 	sch_tree_lock(sch);
 	*old = q->queues[band];
 	q->queues[band] = new;
-	sch->q.qlen -= (*old)->q.qlen;
+	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 	qdisc_reset(*old);
 	sch_tree_unlock(sch);
 
diff --git a/net/sched/sch_red.c b/net/sched/sch_red.c
index d65cadddea6..acddad08850 100644
--- a/net/sched/sch_red.c
+++ b/net/sched/sch_red.c
@@ -175,12 +175,14 @@ static void red_destroy(struct Qdisc *sch)
 	qdisc_destroy(q->qdisc);
 }
 
-static struct Qdisc *red_create_dflt(struct net_device *dev, u32 limit)
+static struct Qdisc *red_create_dflt(struct Qdisc *sch, u32 limit)
 {
-	struct Qdisc *q = qdisc_create_dflt(dev, &bfifo_qdisc_ops);
+	struct Qdisc *q;
 	struct rtattr *rta;
 	int ret;
 
+	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
+			      TC_H_MAKE(sch->handle, 1));
 	if (q) {
 		rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)),
 		              GFP_KERNEL);
@@ -219,7 +221,7 @@ static int red_change(struct Qdisc *sch, struct rtattr *opt)
 	ctl = RTA_DATA(tb[TCA_RED_PARMS-1]);
 
 	if (ctl->limit > 0) {
-		child = red_create_dflt(sch->dev, ctl->limit);
+		child = red_create_dflt(sch, ctl->limit);
 		if (child == NULL)
 			return -ENOMEM;
 	}
@@ -227,8 +229,10 @@ static int red_change(struct Qdisc *sch, struct rtattr *opt)
 	sch_tree_lock(sch);
 	q->flags = ctl->flags;
 	q->limit = ctl->limit;
-	if (child)
+	if (child) {
+		qdisc_tree_decrease_qlen(q->qdisc, q->qdisc->q.qlen);
 		qdisc_destroy(xchg(&q->qdisc, child));
+	}
 
 	red_set_parms(&q->parms, ctl->qth_min, ctl->qth_max, ctl->Wlog,
 				 ctl->Plog, ctl->Scell_log,
@@ -306,8 +310,8 @@ static int red_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	sch_tree_lock(sch);
 	*old = xchg(&q->qdisc, new);
+	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 	qdisc_reset(*old);
-	sch->q.qlen = 0;
 	sch_tree_unlock(sch);
 	return 0;
 }
diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index d0d6e595a78..459cda258a5 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -393,6 +393,7 @@ static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
 {
 	struct sfq_sched_data *q = qdisc_priv(sch);
 	struct tc_sfq_qopt *ctl = RTA_DATA(opt);
+	unsigned int qlen;
 
 	if (opt->rta_len < RTA_LENGTH(sizeof(*ctl)))
 		return -EINVAL;
@@ -403,8 +404,10 @@ static int sfq_change(struct Qdisc *sch, struct rtattr *opt)
 	if (ctl->limit)
 		q->limit = min_t(u32, ctl->limit, SFQ_DEPTH);
 
+	qlen = sch->q.qlen;
 	while (sch->q.qlen >= q->limit-1)
 		sfq_drop(sch);
+	qdisc_tree_decrease_qlen(sch, qlen - sch->q.qlen);
 
 	del_timer(&q->perturb_timer);
 	if (q->perturb_period) {
diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c
index d9a5d298d75..ed9b6d93854 100644
--- a/net/sched/sch_tbf.c
+++ b/net/sched/sch_tbf.c
@@ -250,7 +250,7 @@ static struct sk_buff *tbf_dequeue(struct Qdisc* sch)
 
 		if (q->qdisc->ops->requeue(skb, q->qdisc) != NET_XMIT_SUCCESS) {
 			/* When requeue fails skb is dropped */
-			sch->q.qlen--;
+			qdisc_tree_decrease_qlen(q->qdisc, 1);
 			sch->qstats.drops++;
 		}
 
@@ -273,12 +273,14 @@ static void tbf_reset(struct Qdisc* sch)
 	del_timer(&q->wd_timer);
 }
 
-static struct Qdisc *tbf_create_dflt_qdisc(struct net_device *dev, u32 limit)
+static struct Qdisc *tbf_create_dflt_qdisc(struct Qdisc *sch, u32 limit)
 {
-	struct Qdisc *q = qdisc_create_dflt(dev, &bfifo_qdisc_ops);
+	struct Qdisc *q;
         struct rtattr *rta;
 	int ret;
 
+	q = qdisc_create_dflt(sch->dev, &bfifo_qdisc_ops,
+			      TC_H_MAKE(sch->handle, 1));
 	if (q) {
 		rta = kmalloc(RTA_LENGTH(sizeof(struct tc_fifo_qopt)), GFP_KERNEL);
 		if (rta) {
@@ -341,13 +343,15 @@ static int tbf_change(struct Qdisc* sch, struct rtattr *opt)
 		goto done;
 
 	if (qopt->limit > 0) {
-		if ((child = tbf_create_dflt_qdisc(sch->dev, qopt->limit)) == NULL)
+		if ((child = tbf_create_dflt_qdisc(sch, qopt->limit)) == NULL)
 			goto done;
 	}
 
 	sch_tree_lock(sch);
-	if (child)
+	if (child) {
+		qdisc_tree_decrease_qlen(q->qdisc, q->qdisc->q.qlen);
 		qdisc_destroy(xchg(&q->qdisc, child));
+	}
 	q->limit = qopt->limit;
 	q->mtu = qopt->mtu;
 	q->max_size = max_size;
@@ -449,8 +453,8 @@ static int tbf_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new,
 
 	sch_tree_lock(sch);
 	*old = xchg(&q->qdisc, new);
+	qdisc_tree_decrease_qlen(*old, (*old)->q.qlen);
 	qdisc_reset(*old);
-	sch->q.qlen = 0;
 	sch_tree_unlock(sch);
 
 	return 0;