Net fixes:

- Handle downing of multicast routes properly, since they
refer potentially to multiple devices.
- IPv6 raw had missing sk->filter handling and rawv6_rcv
missing some checksum processing.
- IPV6 udp needs to use skb_checksum not csum_partial on
skb->h.raw
- Ingress packet scheduler fixes from Jamal
- Addrconf bug discoverd by Petr Baudis, ipv6_ifa_notify should
pass NULL second arg to ipv6_chk_addr in RTM_DELADDR
handling.

net/ipv4/fib_frontend.c

@@ -604,6 +604,11 @@ static int fib_netdev_event(struct notifier_block *this, unsigned long event, vo
 	struct net_device *dev = ptr;
 	struct in_device *in_dev = __in_dev_get(dev);
 
+	if (event == NETDEV_UNREGISTER) {
+		fib_disable_ip(dev, 2);
+		return NOTIFY_DONE;
+	}
+
 	if (!in_dev)
 		return NOTIFY_DONE;
 
@@ -620,9 +625,6 @@ static int fib_netdev_event(struct notifier_block *this, unsigned long event, vo
 	case NETDEV_DOWN:
 		fib_disable_ip(dev, 0);
 		break;
-	case NETDEV_UNREGISTER:
-		fib_disable_ip(dev, 1);
-		break;
 	case NETDEV_CHANGEMTU:
 	case NETDEV_CHANGE:
 		rt_cache_flush(0);

net/ipv4/fib_semantics.c

@@ -185,6 +185,7 @@ int ip_fib_check_default(u32 gw, struct net_device *dev)
 			continue;
 		for_nexthops(fi) {
 			if (nh->nh_dev == dev && nh->nh_gw == gw &&
+			    nh->nh_scope == RT_SCOPE_LINK &&
 			    !(nh->nh_flags&RTNH_F_DEAD)) {
 				read_unlock(&fib_info_lock);
 				return 0;
@@ -379,15 +380,23 @@ static int fib_check_nh(const struct rtmsg *r, struct fib_info *fi, struct fib_n
 		/* It is not necessary, but requires a bit of thinking */
 		if (key.scope < RT_SCOPE_LINK)
 			key.scope = RT_SCOPE_LINK;
-
 		if ((err = fib_lookup(&key, &res)) != 0)
 			return err;
+		err = -EINVAL;
+		if (res.type != RTN_UNICAST && res.type != RTN_LOCAL)
+			goto out;
 		nh->nh_scope = res.scope;
 		nh->nh_oif = FIB_RES_OIF(res);
-		nh->nh_dev = FIB_RES_DEV(res);
-		if (nh->nh_dev)
-			atomic_inc(&nh->nh_dev->refcnt);
+		if ((nh->nh_dev = FIB_RES_DEV(res)) == NULL)
+			goto out;
+		atomic_inc(&nh->nh_dev->refcnt);
+		err = -ENETDOWN;
+		if (!(nh->nh_dev->flags & IFF_UP))
+			goto out;
+		err = 0;
+out:
 		fib_res_put(&res);
+		return err;
 	} else {
 		struct in_device *in_dev;
 
@@ -875,13 +884,15 @@ int fib_sync_down(u32 local, struct net_device *dev, int force)
 					fi->fib_power -= nh->nh_power;
 					nh->nh_power = 0;
 					spin_unlock_bh(&fib_multipath_lock);
-					if (force && nh->nh_dev) {
-						dev_put(nh->nh_dev);
-						nh->nh_dev = NULL;
-					}
 #endif
 					dead++;
 				}
+#ifdef CONFIG_IP_ROUTE_MULTIPATH
+				if (force > 1 && nh->nh_dev == dev) {
+					dead = fi->fib_nhs;
+					break;
+				}
+#endif
 			} endfor_nexthops(fi)
 			if (dead == fi->fib_nhs) {
 				fi->fib_flags |= RTNH_F_DEAD;
@@ -914,10 +925,6 @@ int fib_sync_up(struct net_device *dev)
 				alive++;
 				continue;
 			}
-			if (nh->nh_dev == NULL && nh->nh_oif == dev->ifindex) {
-				dev_hold(dev);
-				nh->nh_dev = dev;
-			}
 			if (nh->nh_dev == NULL || !(nh->nh_dev->flags&IFF_UP))
 				continue;
 			if (nh->nh_dev != dev || __in_dev_get(dev) == NULL)

net/ipv6/addrconf.c

@@ -1833,7 +1833,7 @@ static void ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
 		break;
 	case RTM_DELADDR:
 		addrconf_leave_solict(ifp->idev->dev, &ifp->addr);
-		if (!ipv6_chk_addr(&ifp->addr, ifp->idev->dev))
+		if (!ipv6_chk_addr(&ifp->addr, NULL))
 			ip6_rt_addr_del(&ifp->addr, ifp->idev->dev);
 		break;
 	}

net/ipv6/raw.c

@@ -281,6 +281,16 @@ void rawv6_err(struct sock *sk, struct sk_buff *skb,
 
 static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb)
 {
+#if defined(CONFIG_FILTER)
+	if (sk->filter && skb->ip_summed != CHECKSUM_UNNECESSARY) {
+		if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) {
+			IP6_INC_STATS_BH(Ip6InDiscards);
+			kfree_skb(skb);
+			return 0;
+		}
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+	}
+#endif
 	/* Charge it to the socket. */
 	if (sock_queue_rcv_skb(sk,skb)<0) {
 		IP6_INC_STATS_BH(Ip6InDiscards);
@@ -302,10 +312,35 @@ static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb)
 int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
 {
 	struct inet_opt *inet = inet_sk(sk);
+	struct raw6_opt *raw_opt = raw6_sk(sk);
+
+	if (!raw_opt->checksum)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	if (skb->ip_summed != CHECKSUM_UNNECESSARY) {
+		if (skb->ip_summed == CHECKSUM_HW) {
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
+			if (csum_ipv6_magic(&skb->nh.ipv6h->saddr,
+					    &skb->nh.ipv6h->daddr,
+					    skb->len, inet->num, skb->csum)) {
+				NETDEBUG(if (net_ratelimit()) printk(KERN_DEBUG "raw v6 hw csum failure.\n"));
+				skb->ip_summed = CHECKSUM_NONE;
+			}
+		}
+		if (skb->ip_summed == CHECKSUM_NONE)
+			skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr,
+						     &skb->nh.ipv6h->daddr,
+						     skb->len, inet->num, 0);
+	}
 
 	if (inet->hdrincl) {
-		__skb_push(skb, skb->nh.raw - skb->data);
-		skb->h.raw = skb->nh.raw;
+		if (skb->ip_summed != CHECKSUM_UNNECESSARY &&
+		    (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) {
+			IP6_INC_STATS_BH(Ip6InDiscards);
+			kfree_skb(skb);
+			return 0;
+		}
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 
 	rawv6_rcv_skb(sk, skb);
@@ -345,7 +380,17 @@ int rawv6_recvmsg(struct sock *sk, struct msghdr *msg, int len,
   		msg->msg_flags |= MSG_TRUNC;
   	}
 
-	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
+	if (skb->ip_summed==CHECKSUM_UNNECESSARY) {
+		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
+	} else if (msg->msg_flags&MSG_TRUNC) {
+		if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum)))
+			goto csum_copy_err;
+		err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
+	} else {
+		err = skb_copy_and_csum_datagram_iovec(skb, 0, msg->msg_iov);
+		if (err == -EINVAL)
+			goto csum_copy_err;
+	}
 	if (err)
 		goto out_free;
 
@@ -372,6 +417,27 @@ int rawv6_recvmsg(struct sock *sk, struct msghdr *msg, int len,
 	skb_free_datagram(sk, skb);
 out:
 	return err;
+
+csum_copy_err:
+	/* Clear queue. */
+	if (flags&MSG_PEEK) {
+		int clear = 0;
+		spin_lock_irq(&sk->receive_queue.lock);
+		if (skb == skb_peek(&sk->receive_queue)) {
+			__skb_unlink(skb, &sk->receive_queue);
+			clear = 1;
+		}
+		spin_unlock_irq(&sk->receive_queue.lock);
+		if (clear)
+			kfree_skb(skb);
+	}
+
+	/* Error for blocking case is chosen to masquerade
+	   as some normal condition.
+	 */
+	err = (flags&MSG_DONTWAIT) ? -EAGAIN : -EHOSTUNREACH;
+	IP6_INC_STATS_USER(Ip6InDiscards);
+	goto out_free;
 }
 
 /*

net/ipv6/udp.c

@@ -508,7 +508,7 @@ static inline int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
 {
 #if defined(CONFIG_FILTER)
 	if (sk->filter && skb->ip_summed != CHECKSUM_UNNECESSARY) {
-		if ((unsigned short)csum_fold(csum_partial(skb->h.raw, skb->len, skb->csum))) {
+		if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) {
 			UDP6_INC_STATS_BH(UdpInErrors);
 			IP6_INC_STATS_BH(Ip6InDiscards);
 			kfree_skb(skb);

net/sched/sch_ingress.c

@@ -43,6 +43,9 @@
 #define PRIV(sch) ((struct ingress_qdisc_data *) (sch)->data)
 
 
+/* Thanks to Doron Oz for this hack
+*/
+static int nf_registered = 0; 
 
 struct ingress_qdisc_data {
 	struct Qdisc		*q;
@@ -147,15 +150,21 @@ static int ingress_enqueue(struct sk_buff *skb,struct Qdisc *sch)
 #ifdef CONFIG_NET_CLS_POLICE
 		case TC_POLICE_SHOT:
 			result = NF_DROP;
+			sch->stats.drops++;
 			break;
 		case TC_POLICE_RECLASSIFY: /* DSCP remarking here ? */
 		case TC_POLICE_OK:
 		case TC_POLICE_UNSPEC:
 		default:
+			sch->stats.packets++;
+			sch->stats.bytes += skb->len;
 			result = NF_ACCEPT;
 			break;
+	}
+#else
+	sch->stats.packets++;
+	sch->stats.bytes += skb->len;
 #endif
-	};
 
 	skb->tc_index = TC_H_MIN(res.classid);
 	return result;
@@ -236,22 +245,21 @@ int ingress_init(struct Qdisc *sch,struct rtattr *opt)
 {
 	struct ingress_qdisc_data *p = PRIV(sch);
 
+	if (!nf_registered) {
+		if (nf_register_hook(&ing_ops) < 0) {
+			printk("ingress qdisc registration error \n");
+			goto error;
+			}
+		nf_registered++;
+	}
+
 	DPRINTK("ingress_init(sch %p,[qdisc %p],opt %p)\n",sch,p,opt);
 	memset(p, 0, sizeof(*p));
 	p->filter_list = NULL;
 	p->q = &noop_qdisc;
-#ifndef MODULE
-	if (nf_register_hook(&ing_ops) < 0) {
-		printk("Unable to register ingress \n");
-		goto error;
-	}
-#endif
-	DPRINTK("ingress_init: qdisc %p\n", sch);
 	MOD_INC_USE_COUNT;
 	return 0;
-#ifndef MODULE
 error:
-#endif
 	return -EINVAL;
 }
 
@@ -295,12 +303,9 @@ static void ingress_destroy(struct Qdisc *sch)
 /* for future use */
 	qdisc_destroy(p->q);
 #endif
-
-#ifndef MODULE
-	nf_unregister_hook(&ing_ops);
-#endif
-
+ 
 	MOD_DEC_USE_COUNT;
+
 }
 
 
@@ -356,6 +361,7 @@ struct Qdisc_ops ingress_qdisc_ops =
 	ingress_dump,			/* dump */
 };
 
+
 #ifdef MODULE
 int init_module(void)
 {
@@ -366,20 +372,15 @@ int init_module(void)
 		return ret;
 	}
 
-        if (nf_register_hook(&ing_ops) < 0) {
-		printk("Unable to register ingress on hook \n");
-		unregister_qdisc(&ingress_qdisc_ops);
-		return 0;
-	}
-
 	return ret;
 }
 
 
 void cleanup_module(void) 
 {
-	nf_unregister_hook(&ing_ops);
 	unregister_qdisc(&ingress_qdisc_ops);
+	if (nf_registered)
+		nf_unregister_hook(&ing_ops);
 }
 #endif
 MODULE_LICENSE("GPL");