X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=net%2Fipv4%2Fipip.c;h=eda04fed337906fb87be32991fde83f49ff55b25;hb=3705e11a21bcdffe7422ee7870e1b23fe4ac70f4;hp=396437242a1b36186db17ff5aaa1bf5d60a1734b;hpb=cfbba49d80be6cf8d3872b66fc5421f119843b36;p=safe%2Fjmp%2Flinux-2.6

diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 3964372..eda04fe 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -1,8 +1,6 @@
 /*
  *	Linux NET3:	IP/IP protocol decoder.
  *
- *	Version: $Id: ipip.c,v 1.50 2001/10/02 02:22:36 davem Exp $
- *
  *	Authors:
  *		Sam Lantinga (slouken@cs.ucdavis.edu)  02/01/95
  *
@@ -43,7 +41,7 @@
 		Made the tunnels use dev->name not tunnel: when error reporting.
 		Added tx_dropped stat
 
-		-Alan Cox	(Alan.Cox@linux.org) 21 March 95
+		-Alan Cox	(alan@lxorguk.ukuu.org.uk) 21 March 95
 
 	Reworked:
 		Changed to tunnel to destination gateway in addition to the
@@ -115,49 +113,64 @@
 #include <net/ipip.h>
 #include <net/inet_ecn.h>
 #include <net/xfrm.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
 
 #define HASH_SIZE  16
 #define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF)
 
-static int ipip_fb_tunnel_init(struct net_device *dev);
-static int ipip_tunnel_init(struct net_device *dev);
-static void ipip_tunnel_setup(struct net_device *dev);
+static int ipip_net_id __read_mostly;
+struct ipip_net {
+	struct ip_tunnel *tunnels_r_l[HASH_SIZE];
+	struct ip_tunnel *tunnels_r[HASH_SIZE];
+	struct ip_tunnel *tunnels_l[HASH_SIZE];
+	struct ip_tunnel *tunnels_wc[1];
+	struct ip_tunnel **tunnels[4];
 
-static struct net_device *ipip_fb_tunnel_dev;
+	struct net_device *fb_tunnel_dev;
+};
 
-static struct ip_tunnel *tunnels_r_l[HASH_SIZE];
-static struct ip_tunnel *tunnels_r[HASH_SIZE];
-static struct ip_tunnel *tunnels_l[HASH_SIZE];
-static struct ip_tunnel *tunnels_wc[1];
-static struct ip_tunnel **tunnels[4] = { tunnels_wc, tunnels_l, tunnels_r, tunnels_r_l };
+static void ipip_fb_tunnel_init(struct net_device *dev);
+static void ipip_tunnel_init(struct net_device *dev);
+static void ipip_tunnel_setup(struct net_device *dev);
+
+/*
+ * Locking : hash tables are protected by RCU and a spinlock
+ */
+static DEFINE_SPINLOCK(ipip_lock);
 
-static DEFINE_RWLOCK(ipip_lock);
+#define for_each_ip_tunnel_rcu(start) \
+	for (t = rcu_dereference(start); t; t = rcu_dereference(t->next))
 
-static struct ip_tunnel * ipip_tunnel_lookup(__be32 remote, __be32 local)
+static struct ip_tunnel * ipip_tunnel_lookup(struct net *net,
+		__be32 remote, __be32 local)
 {
 	unsigned h0 = HASH(remote);
 	unsigned h1 = HASH(local);
 	struct ip_tunnel *t;
+	struct ipip_net *ipn = net_generic(net, ipip_net_id);
 
-	for (t = tunnels_r_l[h0^h1]; t; t = t->next) {
+	for_each_ip_tunnel_rcu(ipn->tunnels_r_l[h0 ^ h1])
 		if (local == t->parms.iph.saddr &&
 		    remote == t->parms.iph.daddr && (t->dev->flags&IFF_UP))
 			return t;
-	}
-	for (t = tunnels_r[h0]; t; t = t->next) {
+
+	for_each_ip_tunnel_rcu(ipn->tunnels_r[h0])
 		if (remote == t->parms.iph.daddr && (t->dev->flags&IFF_UP))
 			return t;
-	}
-	for (t = tunnels_l[h1]; t; t = t->next) {
+
+	for_each_ip_tunnel_rcu(ipn->tunnels_l[h1])
 		if (local == t->parms.iph.saddr && (t->dev->flags&IFF_UP))
 			return t;
-	}
-	if ((t = tunnels_wc[0]) != NULL && (t->dev->flags&IFF_UP))
+
+	t = rcu_dereference(ipn->tunnels_wc[0]);
+	if (t && (t->dev->flags&IFF_UP))
 		return t;
 	return NULL;
 }
 
-static struct ip_tunnel **__ipip_bucket(struct ip_tunnel_parm *parms)
+static struct ip_tunnel **__ipip_bucket(struct ipip_net *ipn,
+		struct ip_tunnel_parm *parms)
 {
 	__be32 remote = parms->iph.daddr;
 	__be32 local = parms->iph.saddr;
@@ -172,47 +185,50 @@ static struct ip_tunnel **__ipip_bucket(struct ip_tunnel_parm *parms)
 		prio |= 1;
 		h ^= HASH(local);
 	}
-	return &tunnels[prio][h];
+	return &ipn->tunnels[prio][h];
 }
 
-static inline struct ip_tunnel **ipip_bucket(struct ip_tunnel *t)
+static inline struct ip_tunnel **ipip_bucket(struct ipip_net *ipn,
+		struct ip_tunnel *t)
 {
-	return __ipip_bucket(&t->parms);
+	return __ipip_bucket(ipn, &t->parms);
 }
 
-static void ipip_tunnel_unlink(struct ip_tunnel *t)
+static void ipip_tunnel_unlink(struct ipip_net *ipn, struct ip_tunnel *t)
 {
 	struct ip_tunnel **tp;
 
-	for (tp = ipip_bucket(t); *tp; tp = &(*tp)->next) {
+	for (tp = ipip_bucket(ipn, t); *tp; tp = &(*tp)->next) {
 		if (t == *tp) {
-			write_lock_bh(&ipip_lock);
+			spin_lock_bh(&ipip_lock);
 			*tp = t->next;
-			write_unlock_bh(&ipip_lock);
+			spin_unlock_bh(&ipip_lock);
 			break;
 		}
 	}
 }
 
-static void ipip_tunnel_link(struct ip_tunnel *t)
+static void ipip_tunnel_link(struct ipip_net *ipn, struct ip_tunnel *t)
 {
-	struct ip_tunnel **tp = ipip_bucket(t);
+	struct ip_tunnel **tp = ipip_bucket(ipn, t);
 
+	spin_lock_bh(&ipip_lock);
 	t->next = *tp;
-	write_lock_bh(&ipip_lock);
-	*tp = t;
-	write_unlock_bh(&ipip_lock);
+	rcu_assign_pointer(*tp, t);
+	spin_unlock_bh(&ipip_lock);
 }
 
-static struct ip_tunnel * ipip_tunnel_locate(struct ip_tunnel_parm *parms, int create)
+static struct ip_tunnel * ipip_tunnel_locate(struct net *net,
+		struct ip_tunnel_parm *parms, int create)
 {
 	__be32 remote = parms->iph.daddr;
 	__be32 local = parms->iph.saddr;
 	struct ip_tunnel *t, **tp, *nt;
 	struct net_device *dev;
 	char name[IFNAMSIZ];
+	struct ipip_net *ipn = net_generic(net, ipip_net_id);
 
-	for (tp = __ipip_bucket(parms); (t = *tp) != NULL; tp = &t->next) {
+	for (tp = __ipip_bucket(ipn, parms); (t = *tp) != NULL; tp = &t->next) {
 		if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr)
 			return t;
 	}
@@ -221,59 +237,59 @@ static struct ip_tunnel * ipip_tunnel_locate(struct ip_tunnel_parm *parms, int c
 
 	if (parms->name[0])
 		strlcpy(name, parms->name, IFNAMSIZ);
-	else {
-		int i;
-		for (i=1; i<100; i++) {
-			sprintf(name, "tunl%d", i);
-			if (__dev_get_by_name(name) == NULL)
-				break;
-		}
-		if (i==100)
-			goto failed;
-	}
+	else
+		sprintf(name, "tunl%%d");
 
 	dev = alloc_netdev(sizeof(*t), name, ipip_tunnel_setup);
 	if (dev == NULL)
 		return NULL;
 
+	dev_net_set(dev, net);
+
+	if (strchr(name, '%')) {
+		if (dev_alloc_name(dev, name) < 0)
+			goto failed_free;
+	}
+
 	nt = netdev_priv(dev);
-	SET_MODULE_OWNER(dev);
-	dev->init = ipip_tunnel_init;
 	nt->parms = *parms;
 
-	if (register_netdevice(dev) < 0) {
-		free_netdev(dev);
-		goto failed;
-	}
+	ipip_tunnel_init(dev);
+
+	if (register_netdevice(dev) < 0)
+		goto failed_free;
 
 	dev_hold(dev);
-	ipip_tunnel_link(nt);
+	ipip_tunnel_link(ipn, nt);
 	return nt;
 
-failed:
+failed_free:
+	free_netdev(dev);
 	return NULL;
 }
 
 static void ipip_tunnel_uninit(struct net_device *dev)
 {
-	if (dev == ipip_fb_tunnel_dev) {
-		write_lock_bh(&ipip_lock);
-		tunnels_wc[0] = NULL;
-		write_unlock_bh(&ipip_lock);
+	struct net *net = dev_net(dev);
+	struct ipip_net *ipn = net_generic(net, ipip_net_id);
+
+	if (dev == ipn->fb_tunnel_dev) {
+		spin_lock_bh(&ipip_lock);
+		ipn->tunnels_wc[0] = NULL;
+		spin_unlock_bh(&ipip_lock);
 	} else
-		ipip_tunnel_unlink(netdev_priv(dev));
+		ipip_tunnel_unlink(ipn, netdev_priv(dev));
 	dev_put(dev);
 }
 
 static int ipip_err(struct sk_buff *skb, u32 info)
 {
-#ifndef I_WISH_WORLD_WERE_PERFECT
 
-/* It is not :-( All the routers (except for Linux) return only
+/* All the routers (except for Linux) return only
    8 bytes of packet payload. It means, that precise relaying of
    ICMP in the real Internet is absolutely infeasible.
  */
-	struct iphdr *iph = (struct iphdr*)skb->data;
+	struct iphdr *iph = (struct iphdr *)skb->data;
 	const int type = icmp_hdr(skb)->type;
 	const int code = icmp_hdr(skb)->code;
 	struct ip_tunnel *t;
@@ -309,8 +325,8 @@ static int ipip_err(struct sk_buff *skb, u32 info)
 
 	err = -ENOENT;
 
-	read_lock(&ipip_lock);
-	t = ipip_tunnel_lookup(iph->daddr, iph->saddr);
+	rcu_read_lock();
+	t = ipip_tunnel_lookup(dev_net(skb->dev), iph->daddr, iph->saddr);
 	if (t == NULL || t->parms.iph.daddr == 0)
 		goto out;
 
@@ -318,141 +334,14 @@ static int ipip_err(struct sk_buff *skb, u32 info)
 	if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
 		goto out;
 
-	if (jiffies - t->err_time < IPTUNNEL_ERR_TIMEO)
+	if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
 		t->err_count++;
 	else
 		t->err_count = 1;
 	t->err_time = jiffies;
 out:
-	read_unlock(&ipip_lock);
+	rcu_read_unlock();
 	return err;
-#else
-	struct iphdr *iph = (struct iphdr*)dp;
-	int hlen = iph->ihl<<2;
-	struct iphdr *eiph;
-	const int type = icmp_hdr(skb)->type;
-	const int code = icmp_hdr(skb)->code;
-	int rel_type = 0;
-	int rel_code = 0;
-	__be32 rel_info = 0;
-	__u32 n = 0;
-	struct sk_buff *skb2;
-	struct flowi fl;
-	struct rtable *rt;
-
-	if (len < hlen + sizeof(struct iphdr))
-		return 0;
-	eiph = (struct iphdr*)(dp + hlen);
-
-	switch (type) {
-	default:
-		return 0;
-	case ICMP_PARAMETERPROB:
-		n = ntohl(icmp_hdr(skb)->un.gateway) >> 24;
-		if (n < hlen)
-			return 0;
-
-		/* So... This guy found something strange INSIDE encapsulated
-		   packet. Well, he is fool, but what can we do ?
-		 */
-		rel_type = ICMP_PARAMETERPROB;
-		rel_info = htonl((n - hlen) << 24);
-		break;
-
-	case ICMP_DEST_UNREACH:
-		switch (code) {
-		case ICMP_SR_FAILED:
-		case ICMP_PORT_UNREACH:
-			/* Impossible event. */
-			return 0;
-		case ICMP_FRAG_NEEDED:
-			/* And it is the only really necessary thing :-) */
-			n = ntohs(icmp_hdr(skb)->un.frag.mtu);
-			if (n < hlen+68)
-				return 0;
-			n -= hlen;
-			/* BSD 4.2 MORE DOES NOT EXIST IN NATURE. */
-			if (n > ntohs(eiph->tot_len))
-				return 0;
-			rel_info = htonl(n);
-			break;
-		default:
-			/* All others are translated to HOST_UNREACH.
-			   rfc2003 contains "deep thoughts" about NET_UNREACH,
-			   I believe, it is just ether pollution. --ANK
-			 */
-			rel_type = ICMP_DEST_UNREACH;
-			rel_code = ICMP_HOST_UNREACH;
-			break;
-		}
-		break;
-	case ICMP_TIME_EXCEEDED:
-		if (code != ICMP_EXC_TTL)
-			return 0;
-		break;
-	}
-
-	/* Prepare fake skb to feed it to icmp_send */
-	skb2 = skb_clone(skb, GFP_ATOMIC);
-	if (skb2 == NULL)
-		return 0;
-	dst_release(skb2->dst);
-	skb2->dst = NULL;
-	skb_pull(skb2, skb->data - (u8*)eiph);
-	skb_reset_network_header(skb2);
-
-	/* Try to guess incoming interface */
-	memset(&fl, 0, sizeof(fl));
-	fl.fl4_daddr = eiph->saddr;
-	fl.fl4_tos = RT_TOS(eiph->tos);
-	fl.proto = IPPROTO_IPIP;
-	if (ip_route_output_key(&rt, &key)) {
-		kfree_skb(skb2);
-		return 0;
-	}
-	skb2->dev = rt->u.dst.dev;
-
-	/* route "incoming" packet */
-	if (rt->rt_flags&RTCF_LOCAL) {
-		ip_rt_put(rt);
-		rt = NULL;
-		fl.fl4_daddr = eiph->daddr;
-		fl.fl4_src = eiph->saddr;
-		fl.fl4_tos = eiph->tos;
-		if (ip_route_output_key(&rt, &fl) ||
-		    rt->u.dst.dev->type != ARPHRD_TUNNEL) {
-			ip_rt_put(rt);
-			kfree_skb(skb2);
-			return 0;
-		}
-	} else {
-		ip_rt_put(rt);
-		if (ip_route_input(skb2, eiph->daddr, eiph->saddr, eiph->tos, skb2->dev) ||
-		    skb2->dst->dev->type != ARPHRD_TUNNEL) {
-			kfree_skb(skb2);
-			return 0;
-		}
-	}
-
-	/* change mtu on this route */
-	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
-		if (n > dst_mtu(skb2->dst)) {
-			kfree_skb(skb2);
-			return 0;
-		}
-		skb2->dst->ops->update_pmtu(skb2->dst, n);
-	} else if (type == ICMP_TIME_EXCEEDED) {
-		struct ip_tunnel *t = netdev_priv(skb2->dev);
-		if (t->parms.iph.ttl) {
-			rel_type = ICMP_DEST_UNREACH;
-			rel_code = ICMP_HOST_UNREACH;
-		}
-	}
-
-	icmp_send(skb2, rel_type, rel_code, rel_info);
-	kfree_skb(skb2);
-	return 0;
-#endif
 }
 
 static inline void ipip_ecn_decapsulate(const struct iphdr *outer_iph,
@@ -469,10 +358,11 @@ static int ipip_rcv(struct sk_buff *skb)
 	struct ip_tunnel *tunnel;
 	const struct iphdr *iph = ip_hdr(skb);
 
-	read_lock(&ipip_lock);
-	if ((tunnel = ipip_tunnel_lookup(iph->saddr, iph->daddr)) != NULL) {
+	rcu_read_lock();
+	if ((tunnel = ipip_tunnel_lookup(dev_net(skb->dev),
+					iph->saddr, iph->daddr)) != NULL) {
 		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
-			read_unlock(&ipip_lock);
+			rcu_read_unlock();
 			kfree_skb(skb);
 			return 0;
 		}
@@ -484,18 +374,17 @@ static int ipip_rcv(struct sk_buff *skb)
 		skb->protocol = htons(ETH_P_IP);
 		skb->pkt_type = PACKET_HOST;
 
-		tunnel->stat.rx_packets++;
-		tunnel->stat.rx_bytes += skb->len;
+		tunnel->dev->stats.rx_packets++;
+		tunnel->dev->stats.rx_bytes += skb->len;
 		skb->dev = tunnel->dev;
-		dst_release(skb->dst);
-		skb->dst = NULL;
+		skb_dst_drop(skb);
 		nf_reset(skb);
 		ipip_ecn_decapsulate(iph, skb);
 		netif_rx(skb);
-		read_unlock(&ipip_lock);
+		rcu_read_unlock();
 		return 0;
 	}
-	read_unlock(&ipip_lock);
+	rcu_read_unlock();
 
 	return -1;
 }
@@ -505,10 +394,11 @@ static int ipip_rcv(struct sk_buff *skb)
  *	and that skb is filled properly by that function.
  */
 
-static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
-	struct net_device_stats *stats = &tunnel->stat;
+	struct net_device_stats *stats = &dev->stats;
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
 	struct iphdr  *tiph = &tunnel->parms.iph;
 	u8     tos = tunnel->parms.iph.tos;
 	__be16 df = tiph->frag_off;
@@ -516,15 +406,10 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	struct net_device *tdev;			/* Device to other host */
 	struct iphdr  *old_iph = ip_hdr(skb);
 	struct iphdr  *iph;			/* Our new IP header */
-	int    max_headroom;			/* The extra header space needed */
+	unsigned int max_headroom;		/* The extra header space needed */
 	__be32 dst = tiph->daddr;
 	int    mtu;
 
-	if (tunnel->recursion++) {
-		tunnel->stat.collisions++;
-		goto tx_error;
-	}
-
 	if (skb->protocol != htons(ETH_P_IP))
 		goto tx_error;
 
@@ -533,8 +418,8 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (!dst) {
 		/* NBMA tunnel */
-		if ((rt = (struct rtable*)skb->dst) == NULL) {
-			tunnel->stat.tx_fifo_errors++;
+		if ((rt = skb_rtable(skb)) == NULL) {
+			stats->tx_fifo_errors++;
 			goto tx_error;
 		}
 		if ((dst = rt->rt_gateway) == 0)
@@ -548,8 +433,8 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 						.saddr = tiph->saddr,
 						.tos = RT_TOS(tos) } },
 				    .proto = IPPROTO_IPIP };
-		if (ip_route_output_key(&rt, &fl)) {
-			tunnel->stat.tx_carrier_errors++;
+		if (ip_route_output_key(dev_net(dev), &rt, &fl)) {
+			stats->tx_carrier_errors++;
 			goto tx_error_icmp;
 		}
 	}
@@ -557,33 +442,36 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (tdev == dev) {
 		ip_rt_put(rt);
-		tunnel->stat.collisions++;
+		stats->collisions++;
 		goto tx_error;
 	}
 
-	if (tiph->frag_off)
+	df |= old_iph->frag_off & htons(IP_DF);
+
+	if (df) {
 		mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr);
-	else
-		mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu;
 
-	if (mtu < 68) {
-		tunnel->stat.collisions++;
-		ip_rt_put(rt);
-		goto tx_error;
-	}
-	if (skb->dst)
-		skb->dst->ops->update_pmtu(skb->dst, mtu);
+		if (mtu < 68) {
+			stats->collisions++;
+			ip_rt_put(rt);
+			goto tx_error;
+		}
 
-	df |= (old_iph->frag_off&htons(IP_DF));
+		if (skb_dst(skb))
+			skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu);
 
-	if ((old_iph->frag_off&htons(IP_DF)) && mtu < ntohs(old_iph->tot_len)) {
-		icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu));
-		ip_rt_put(rt);
-		goto tx_error;
+		if ((old_iph->frag_off & htons(IP_DF)) &&
+		    mtu < ntohs(old_iph->tot_len)) {
+			icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
+				  htonl(mtu));
+			ip_rt_put(rt);
+			goto tx_error;
+		}
 	}
 
 	if (tunnel->err_count > 0) {
-		if (jiffies - tunnel->err_time < IPTUNNEL_ERR_TIMEO) {
+		if (time_before(jiffies,
+				tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
 			tunnel->err_count--;
 			dst_link_failure(skb);
 		} else
@@ -600,10 +488,9 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 		struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
 		if (!new_skb) {
 			ip_rt_put(rt);
-			stats->tx_dropped++;
+			txq->tx_dropped++;
 			dev_kfree_skb(skb);
-			tunnel->recursion--;
-			return 0;
+			return NETDEV_TX_OK;
 		}
 		if (skb->sk)
 			skb_set_owner_w(new_skb, skb->sk);
@@ -618,8 +505,8 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
 	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
 			      IPSKB_REROUTED);
-	dst_release(skb->dst);
-	skb->dst = &rt->u.dst;
+	skb_dst_drop(skb);
+	skb_dst_set(skb, &rt->u.dst);
 
 	/*
 	 *	Push down and install the IPIP header.
@@ -640,16 +527,48 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	nf_reset(skb);
 
 	IPTUNNEL_XMIT();
-	tunnel->recursion--;
-	return 0;
+	return NETDEV_TX_OK;
 
 tx_error_icmp:
 	dst_link_failure(skb);
 tx_error:
 	stats->tx_errors++;
 	dev_kfree_skb(skb);
-	tunnel->recursion--;
-	return 0;
+	return NETDEV_TX_OK;
+}
+
+static void ipip_tunnel_bind_dev(struct net_device *dev)
+{
+	struct net_device *tdev = NULL;
+	struct ip_tunnel *tunnel;
+	struct iphdr *iph;
+
+	tunnel = netdev_priv(dev);
+	iph = &tunnel->parms.iph;
+
+	if (iph->daddr) {
+		struct flowi fl = { .oif = tunnel->parms.link,
+				    .nl_u = { .ip4_u =
+					      { .daddr = iph->daddr,
+						.saddr = iph->saddr,
+						.tos = RT_TOS(iph->tos) } },
+				    .proto = IPPROTO_IPIP };
+		struct rtable *rt;
+		if (!ip_route_output_key(dev_net(dev), &rt, &fl)) {
+			tdev = rt->u.dst.dev;
+			ip_rt_put(rt);
+		}
+		dev->flags |= IFF_POINTOPOINT;
+	}
+
+	if (!tdev && tunnel->parms.link)
+		tdev = __dev_get_by_index(dev_net(dev), tunnel->parms.link);
+
+	if (tdev) {
+		dev->hard_header_len = tdev->hard_header_len + sizeof(struct iphdr);
+		dev->mtu = tdev->mtu - sizeof(struct iphdr);
+	}
+	dev->iflink = tunnel->parms.link;
 }
 
 static int
@@ -658,16 +577,18 @@ ipip_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 	int err = 0;
 	struct ip_tunnel_parm p;
 	struct ip_tunnel *t;
+	struct net *net = dev_net(dev);
+	struct ipip_net *ipn = net_generic(net, ipip_net_id);
 
 	switch (cmd) {
 	case SIOCGETTUNNEL:
 		t = NULL;
-		if (dev == ipip_fb_tunnel_dev) {
+		if (dev == ipn->fb_tunnel_dev) {
 			if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) {
 				err = -EFAULT;
 				break;
 			}
-			t = ipip_tunnel_locate(&p, 0);
+			t = ipip_tunnel_locate(net, &p, 0);
 		}
 		if (t == NULL)
 			t = netdev_priv(dev);
@@ -693,9 +614,9 @@ ipip_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (p.iph.ttl)
 			p.iph.frag_off |= htons(IP_DF);
 
-		t = ipip_tunnel_locate(&p, cmd == SIOCADDTUNNEL);
+		t = ipip_tunnel_locate(net, &p, cmd == SIOCADDTUNNEL);
 
-		if (dev != ipip_fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
+		if (dev != ipn->fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
 			if (t != NULL) {
 				if (t->dev != dev) {
 					err = -EEXIST;
@@ -708,12 +629,12 @@ ipip_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 					break;
 				}
 				t = netdev_priv(dev);
-				ipip_tunnel_unlink(t);
+				ipip_tunnel_unlink(ipn, t);
 				t->parms.iph.saddr = p.iph.saddr;
 				t->parms.iph.daddr = p.iph.daddr;
 				memcpy(dev->dev_addr, &p.iph.saddr, 4);
 				memcpy(dev->broadcast, &p.iph.daddr, 4);
-				ipip_tunnel_link(t);
+				ipip_tunnel_link(ipn, t);
 				netdev_state_change(dev);
 			}
 		}
@@ -724,6 +645,11 @@ ipip_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 				t->parms.iph.ttl = p.iph.ttl;
 				t->parms.iph.tos = p.iph.tos;
 				t->parms.iph.frag_off = p.iph.frag_off;
+				if (t->parms.link != p.link) {
+					t->parms.link = p.link;
+					ipip_tunnel_bind_dev(dev);
+					netdev_state_change(dev);
+				}
 			}
 			if (copy_to_user(ifr->ifr_ifru.ifru_data, &t->parms, sizeof(p)))
 				err = -EFAULT;
@@ -736,15 +662,15 @@ ipip_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
 		if (!capable(CAP_NET_ADMIN))
 			goto done;
 
-		if (dev == ipip_fb_tunnel_dev) {
+		if (dev == ipn->fb_tunnel_dev) {
 			err = -EFAULT;
 			if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
 				goto done;
 			err = -ENOENT;
-			if ((t = ipip_tunnel_locate(&p, 0)) == NULL)
+			if ((t = ipip_tunnel_locate(net, &p, 0)) == NULL)
 				goto done;
 			err = -EPERM;
-			if (t->dev == ipip_fb_tunnel_dev)
+			if (t->dev == ipn->fb_tunnel_dev)
 				goto done;
 			dev = t->dev;
 		}
@@ -760,11 +686,6 @@ done:
 	return err;
 }
 
-static struct net_device_stats *ipip_tunnel_get_stats(struct net_device *dev)
-{
-	return &(((struct ip_tunnel*)netdev_priv(dev))->stat);
-}
-
 static int ipip_tunnel_change_mtu(struct net_device *dev, int new_mtu)
 {
 	if (new_mtu < 68 || new_mtu > 0xFFF8 - sizeof(struct iphdr))
@@ -773,14 +694,17 @@ static int ipip_tunnel_change_mtu(struct net_device *dev, int new_mtu)
 	return 0;
 }
 
+static const struct net_device_ops ipip_netdev_ops = {
+	.ndo_uninit	= ipip_tunnel_uninit,
+	.ndo_start_xmit	= ipip_tunnel_xmit,
+	.ndo_do_ioctl	= ipip_tunnel_ioctl,
+	.ndo_change_mtu	= ipip_tunnel_change_mtu,
+
+};
+
 static void ipip_tunnel_setup(struct net_device *dev)
 {
-	SET_MODULE_OWNER(dev);
-	dev->uninit		= ipip_tunnel_uninit;
-	dev->hard_start_xmit	= ipip_tunnel_xmit;
-	dev->get_stats		= ipip_tunnel_get_stats;
-	dev->do_ioctl		= ipip_tunnel_ioctl;
-	dev->change_mtu		= ipip_tunnel_change_mtu;
+	dev->netdev_ops		= &ipip_netdev_ops;
 	dev->destructor		= free_netdev;
 
 	dev->type		= ARPHRD_TUNNEL;
@@ -789,16 +713,13 @@ static void ipip_tunnel_setup(struct net_device *dev)
 	dev->flags		= IFF_NOARP;
 	dev->iflink		= 0;
 	dev->addr_len		= 4;
+	dev->features		|= NETIF_F_NETNS_LOCAL;
+	dev->priv_flags		&= ~IFF_XMIT_DST_RELEASE;
 }
 
-static int ipip_tunnel_init(struct net_device *dev)
+static void ipip_tunnel_init(struct net_device *dev)
 {
-	struct net_device *tdev = NULL;
-	struct ip_tunnel *tunnel;
-	struct iphdr *iph;
-
-	tunnel = netdev_priv(dev);
-	iph = &tunnel->parms.iph;
+	struct ip_tunnel *tunnel = netdev_priv(dev);
 
 	tunnel->dev = dev;
 	strcpy(tunnel->parms.name, dev->name);
@@ -806,37 +727,14 @@ static int ipip_tunnel_init(struct net_device *dev)
 	memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
 	memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);
 
-	if (iph->daddr) {
-		struct flowi fl = { .oif = tunnel->parms.link,
-				    .nl_u = { .ip4_u =
-					      { .daddr = iph->daddr,
-						.saddr = iph->saddr,
-						.tos = RT_TOS(iph->tos) } },
-				    .proto = IPPROTO_IPIP };
-		struct rtable *rt;
-		if (!ip_route_output_key(&rt, &fl)) {
-			tdev = rt->u.dst.dev;
-			ip_rt_put(rt);
-		}
-		dev->flags |= IFF_POINTOPOINT;
-	}
-
-	if (!tdev && tunnel->parms.link)
-		tdev = __dev_get_by_index(tunnel->parms.link);
-
-	if (tdev) {
-		dev->hard_header_len = tdev->hard_header_len + sizeof(struct iphdr);
-		dev->mtu = tdev->mtu - sizeof(struct iphdr);
-	}
-	dev->iflink = tunnel->parms.link;
-
-	return 0;
+	ipip_tunnel_bind_dev(dev);
 }
 
-static int __init ipip_fb_tunnel_init(struct net_device *dev)
+static void ipip_fb_tunnel_init(struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	struct iphdr *iph = &tunnel->parms.iph;
+	struct ipip_net *ipn = net_generic(dev_net(dev), ipip_net_id);
 
 	tunnel->dev = dev;
 	strcpy(tunnel->parms.name, dev->name);
@@ -846,8 +744,7 @@ static int __init ipip_fb_tunnel_init(struct net_device *dev)
 	iph->ihl		= 5;
 
 	dev_hold(dev);
-	tunnels_wc[0]		= tunnel;
-	return 0;
+	ipn->tunnels_wc[0]	= tunnel;
 }
 
 static struct xfrm_tunnel ipip_handler = {
@@ -856,53 +753,94 @@ static struct xfrm_tunnel ipip_handler = {
 	.priority	=	1,
 };
 
-static char banner[] __initdata =
+static const char banner[] __initconst =
 	KERN_INFO "IPv4 over IPv4 tunneling driver\n";
 
-static int __init ipip_init(void)
+static void ipip_destroy_tunnels(struct ipip_net *ipn, struct list_head *head)
 {
-	int err;
+	int prio;
 
-	printk(banner);
+	for (prio = 1; prio < 4; prio++) {
+		int h;
+		for (h = 0; h < HASH_SIZE; h++) {
+			struct ip_tunnel *t = ipn->tunnels[prio][h];
 
-	if (xfrm4_tunnel_register(&ipip_handler, AF_INET)) {
-		printk(KERN_INFO "ipip init: can't register tunnel\n");
-		return -EAGAIN;
+			while (t != NULL) {
+				unregister_netdevice_queue(t->dev, head);
+				t = t->next;
+			}
+		}
 	}
+}
 
-	ipip_fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel),
+static int ipip_init_net(struct net *net)
+{
+	struct ipip_net *ipn = net_generic(net, ipip_net_id);
+	int err;
+
+	ipn->tunnels[0] = ipn->tunnels_wc;
+	ipn->tunnels[1] = ipn->tunnels_l;
+	ipn->tunnels[2] = ipn->tunnels_r;
+	ipn->tunnels[3] = ipn->tunnels_r_l;
+
+	ipn->fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel),
 					   "tunl0",
 					   ipip_tunnel_setup);
-	if (!ipip_fb_tunnel_dev) {
+	if (!ipn->fb_tunnel_dev) {
 		err = -ENOMEM;
-		goto err1;
+		goto err_alloc_dev;
 	}
+	dev_net_set(ipn->fb_tunnel_dev, net);
+
+	ipip_fb_tunnel_init(ipn->fb_tunnel_dev);
+
+	if ((err = register_netdev(ipn->fb_tunnel_dev)))
+		goto err_reg_dev;
 
-	ipip_fb_tunnel_dev->init = ipip_fb_tunnel_init;
+	return 0;
 
-	if ((err = register_netdev(ipip_fb_tunnel_dev)))
-		goto err2;
- out:
+err_reg_dev:
+	free_netdev(ipn->fb_tunnel_dev);
+err_alloc_dev:
+	/* nothing */
 	return err;
- err2:
-	free_netdev(ipip_fb_tunnel_dev);
- err1:
-	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
-	goto out;
 }
 
-static void __exit ipip_destroy_tunnels(void)
+static void ipip_exit_net(struct net *net)
 {
-	int prio;
+	struct ipip_net *ipn = net_generic(net, ipip_net_id);
+	LIST_HEAD(list);
 
-	for (prio = 1; prio < 4; prio++) {
-		int h;
-		for (h = 0; h < HASH_SIZE; h++) {
-			struct ip_tunnel *t;
-			while ((t = tunnels[prio][h]) != NULL)
-				unregister_netdevice(t->dev);
-		}
+	rtnl_lock();
+	ipip_destroy_tunnels(ipn, &list);
+	unregister_netdevice_queue(ipn->fb_tunnel_dev, &list);
+	unregister_netdevice_many(&list);
+	rtnl_unlock();
+}
+
+static struct pernet_operations ipip_net_ops = {
+	.init = ipip_init_net,
+	.exit = ipip_exit_net,
+	.id   = &ipip_net_id,
+	.size = sizeof(struct ipip_net),
+};
+
+static int __init ipip_init(void)
+{
+	int err;
+
+	printk(banner);
+
+	if (xfrm4_tunnel_register(&ipip_handler, AF_INET)) {
+		printk(KERN_INFO "ipip init: can't register tunnel\n");
+		return -EAGAIN;
 	}
+
+	err = register_pernet_device(&ipip_net_ops);
+	if (err)
+		xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
+
+	return err;
 }
 
 static void __exit ipip_fini(void)
@@ -910,10 +848,7 @@ static void __exit ipip_fini(void)
 	if (xfrm4_tunnel_deregister(&ipip_handler, AF_INET))
 		printk(KERN_INFO "ipip close: can't deregister tunnel\n");
 
-	rtnl_lock();
-	ipip_destroy_tunnels();
-	unregister_netdevice(ipip_fb_tunnel_dev);
-	rtnl_unlock();
+	unregister_pernet_device(&ipip_net_ops);
 }
 
 module_init(ipip_init);