Merge git://git.skbuff.net/gitroot/yoshfuji/linux-2.6.14+git+ipv6-fix-20051221a

include/linux/ipv6_route.h

@@ -18,6 +18,7 @@
 					   fallback, no routers on link */
 #define RTF_ADDRCONF	0x00040000	/* addrconf route - RA		*/
 #define RTF_PREFIX_RT	0x00080000	/* A prefix only route - RA	*/
+#define RTF_ANYCAST	0x00100000	/* Anycast			*/
 
 #define RTF_NONEXTHOP	0x00200000	/* route with no nexthop	*/
 #define RTF_EXPIRES	0x00400000

include/net/if_inet6.h

@@ -24,6 +24,7 @@
 #define IF_RA_MANAGED	0x40
 #define IF_RA_RCVD	0x20
 #define IF_RS_SENT	0x10
+#define IF_READY	0x80000000
 
 /* prefix flags */
 #define IF_PREFIX_ONLINK	0x01

net/ipv6/addrconf.c

@@ -137,6 +137,7 @@ static int addrconf_ifdown(struct net_device *dev, int how);
 static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags);
 static void addrconf_dad_timer(unsigned long data);
 static void addrconf_dad_completed(struct inet6_ifaddr *ifp);
+static void addrconf_dad_run(struct inet6_dev *idev);
 static void addrconf_rs_timer(unsigned long data);
 static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifa);
 static void ipv6_ifa_notify(int event, struct inet6_ifaddr *ifa);
@@ -388,6 +389,9 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev)
 		}
 #endif
 
+		if (netif_carrier_ok(dev))
+			ndev->if_flags |= IF_READY;
+
 		write_lock_bh(&addrconf_lock);
 		dev->ip6_ptr = ndev;
 		write_unlock_bh(&addrconf_lock);
@@ -415,6 +419,7 @@ static struct inet6_dev * ipv6_find_idev(struct net_device *dev)
 		if ((idev = ipv6_add_dev(dev)) == NULL)
 			return NULL;
 	}
+
 	if (dev->flags&IFF_UP)
 		ipv6_mc_up(idev);
 	return idev;
@@ -904,11 +909,18 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev,
 
 			score.addr_type = __ipv6_addr_type(&ifa->addr);
 
-			/* Rule 0: Candidate Source Address (section 4)
+			/* Rule 0:
+			 * - Tentative Address (RFC2462 section 5.4)
+			 *  - A tentative address is not considered
+			 *    "assigned to an interface" in the traditional
+			 *    sense.
+			 * - Candidate Source Address (section 4)
 			 *  - In any case, anycast addresses, multicast
 			 *    addresses, and the unspecified address MUST
 			 *    NOT be included in a candidate set.
 			 */
+			if (ifa->flags & IFA_F_TENTATIVE)
+				continue;
 			if (unlikely(score.addr_type == IPV6_ADDR_ANY ||
 				     score.addr_type & IPV6_ADDR_MULTICAST)) {
 				LIMIT_NETDEBUG(KERN_DEBUG
@@ -1216,10 +1228,8 @@ int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2)
 
 /* Gets referenced address, destroys ifaddr */
 
-void addrconf_dad_failure(struct inet6_ifaddr *ifp)
+void addrconf_dad_stop(struct inet6_ifaddr *ifp)
 {
-	if (net_ratelimit())
-		printk(KERN_INFO "%s: duplicate address detected!\n", ifp->idev->dev->name);
 	if (ifp->flags&IFA_F_PERMANENT) {
 		spin_lock_bh(&ifp->lock);
 		addrconf_del_timer(ifp);
@@ -1245,6 +1255,12 @@ void addrconf_dad_failure(struct inet6_ifaddr *ifp)
 		ipv6_del_addr(ifp);
 }
 
+void addrconf_dad_failure(struct inet6_ifaddr *ifp)
+{
+	if (net_ratelimit())
+		printk(KERN_INFO "%s: duplicate address detected!\n", ifp->idev->dev->name);
+	addrconf_dad_stop(ifp);
+}
 
 /* Join to solicited addr multicast group. */
 
@@ -2134,9 +2150,42 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event,
 {
 	struct net_device *dev = (struct net_device *) data;
 	struct inet6_dev *idev = __in6_dev_get(dev);
+	int run_pending = 0;
 
 	switch(event) {
 	case NETDEV_UP:
+	case NETDEV_CHANGE:
+		if (event == NETDEV_UP) {
+			if (!netif_carrier_ok(dev)) {
+				/* device is not ready yet. */
+				printk(KERN_INFO
+					"ADDRCONF(NETDEV_UP): %s: "
+					"link is not ready\n",
+					dev->name);
+				break;
+			}
+		} else {
+			if (!netif_carrier_ok(dev)) {
+				/* device is still not ready. */
+				break;
+			}
+
+			if (idev) {
+				if (idev->if_flags & IF_READY) {
+					/* device is already configured. */
+					break;
+				}
+				idev->if_flags |= IF_READY;
+			}
+
+			printk(KERN_INFO
+					"ADDRCONF(NETDEV_CHANGE): %s: "
+					"link becomes ready\n",
+					dev->name);
+
+			run_pending = 1;
+		}
+
 		switch(dev->type) {
 		case ARPHRD_SIT:
 			addrconf_sit_config(dev);
@@ -2153,6 +2202,9 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event,
 			break;
 		};
 		if (idev) {
+			if (run_pending)
+				addrconf_dad_run(idev);
+
 			/* If the MTU changed during the interface down, when the
 			   interface up, the changed MTU must be reflected in the
 			   idev as well as routers.
@@ -2187,8 +2239,7 @@ static int addrconf_notify(struct notifier_block *this, unsigned long event,
 		 */
 		addrconf_ifdown(dev, event != NETDEV_DOWN);
 		break;
-	case NETDEV_CHANGE:
-		break;
+
 	case NETDEV_CHANGENAME:
 #ifdef CONFIG_SYSCTL
 		if (idev) {
@@ -2269,7 +2320,7 @@ static int addrconf_ifdown(struct net_device *dev, int how)
 
 	/* Step 3: clear flags for stateless addrconf */
 	if (how != 1)
-		idev->if_flags &= ~(IF_RS_SENT|IF_RA_RCVD);
+		idev->if_flags &= ~(IF_RS_SENT|IF_RA_RCVD|IF_READY);
 
 	/* Step 4: clear address list */
 #ifdef CONFIG_IPV6_PRIVACY
@@ -2378,11 +2429,20 @@ static void addrconf_rs_timer(unsigned long data)
 /*
  *	Duplicate Address Detection
  */
+static void addrconf_dad_kick(struct inet6_ifaddr *ifp)
+{
+	unsigned long rand_num;
+	struct inet6_dev *idev = ifp->idev;
+
+	rand_num = net_random() % (idev->cnf.rtr_solicit_delay ? : 1);
+	ifp->probes = idev->cnf.dad_transmits;
+	addrconf_mod_timer(ifp, AC_DAD, rand_num);
+}
+
 static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags)
 {
 	struct inet6_dev *idev = ifp->idev;
 	struct net_device *dev = idev->dev;
-	unsigned long rand_num;
 
 	addrconf_join_solict(dev, &ifp->addr);
 
@@ -2391,7 +2451,6 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags)
 					flags);
 
 	net_srandom(ifp->addr.s6_addr32[3]);
-	rand_num = net_random() % (idev->cnf.rtr_solicit_delay ? : 1);
 
 	read_lock_bh(&idev->lock);
 	if (ifp->dead)
@@ -2408,8 +2467,17 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags)
 		return;
 	}
 
-	ifp->probes = idev->cnf.dad_transmits;
-	addrconf_mod_timer(ifp, AC_DAD, rand_num);
+	if (idev->if_flags & IF_READY)
+		addrconf_dad_kick(ifp);
+	else {
+		/*
+		 * If the defice is not ready:
+		 * - keep it tentative if it is a permanent address.
+		 * - otherwise, kill it.
+		 */
+		in6_ifa_hold(ifp);
+		addrconf_dad_stop(ifp);
+	}
 
 	spin_unlock_bh(&ifp->lock);
 out:
@@ -2493,6 +2561,22 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp)
 	}
 }
 
+static void addrconf_dad_run(struct inet6_dev *idev) {
+	struct inet6_ifaddr *ifp;
+
+	read_lock_bh(&idev->lock);
+	for (ifp = idev->addr_list; ifp; ifp = ifp->if_next) {
+		spin_lock_bh(&ifp->lock);
+		if (!(ifp->flags & IFA_F_TENTATIVE)) {
+			spin_unlock_bh(&ifp->lock);
+			continue;
+		}
+		spin_unlock_bh(&ifp->lock);
+		addrconf_dad_kick(ifp);
+	}
+	read_unlock_bh(&idev->lock);
+}
+
 #ifdef CONFIG_PROC_FS
 struct if6_iter_state {
 	int bucket;

net/ipv6/icmp.c

@@ -328,8 +328,10 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 		iif = skb->dev->ifindex;
 
 	/*
-	 *	Must not send if we know that source is Anycast also.
-	 *	for now we don't know that.
+	 *	Must not send error if the source does not uniquely
+	 *	identify a single node (RFC2463 Section 2.4).
+	 *	We check unspecified / multicast addresses here,
+	 *	and anycast addresses will be checked later.
 	 */
 	if ((addr_type == IPV6_ADDR_ANY) || (addr_type & IPV6_ADDR_MULTICAST)) {
 		LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: addr_any/mcast source\n");
@@ -373,6 +375,16 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 	err = ip6_dst_lookup(sk, &dst, &fl);
 	if (err)
 		goto out;
+
+	/*
+	 * We won't send icmp if the destination is known
+	 * anycast.
+	 */
+	if (((struct rt6_info *)dst)->rt6i_flags & RTF_ANYCAST) {
+		LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: acast source\n");
+		goto out_dst_release;
+	}
+
 	if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0)
 		goto out;
 

net/ipv6/route.c

@@ -413,11 +413,14 @@ static struct rt6_info *rt6_cow(struct rt6_info *ort, struct in6_addr *daddr,
 	rt = ip6_rt_copy(ort);
 
 	if (rt) {
-		ipv6_addr_copy(&rt->rt6i_dst.addr, daddr);
-
-		if (!(rt->rt6i_flags&RTF_GATEWAY))
+		if (!(rt->rt6i_flags&RTF_GATEWAY)) {
+			if (rt->rt6i_dst.plen != 128 &&
+			    ipv6_addr_equal(&rt->rt6i_dst.addr, daddr))
+				rt->rt6i_flags |= RTF_ANYCAST;
 			ipv6_addr_copy(&rt->rt6i_gateway, daddr);
+		}
 
+		ipv6_addr_copy(&rt->rt6i_dst.addr, daddr);
 		rt->rt6i_dst.plen = 128;
 		rt->rt6i_flags |= RTF_CACHE;
 		rt->u.dst.flags |= DST_HOST;
@@ -1413,7 +1416,9 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev,
 	rt->u.dst.obsolete = -1;
 
 	rt->rt6i_flags = RTF_UP | RTF_NONEXTHOP;
-	if (!anycast)
+	if (anycast)
+		rt->rt6i_flags |= RTF_ANYCAST;
+	else
 		rt->rt6i_flags |= RTF_LOCAL;
 	rt->rt6i_nexthop = ndisc_get_neigh(rt->rt6i_dev, &rt->rt6i_gateway);
 	if (rt->rt6i_nexthop == NULL) {