X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;ds=sidebyside;f=net%2Fipv4%2Fip_gre.c;h=7631b20490f52d34dce1686ff48f2391812129b7;hb=d088dde7b19628f386c486f16cd471f5b5682ca8;hp=4a43739c9035df34a6ec0344662a2d8dab9c6325;hpb=749c10f931923451a4c59b4435d182aa9ae27a4f;p=safe%2Fjmp%2Flinux-2.6

diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 4a43739..7631b20 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -66,10 +66,7 @@
    solution, but it supposes maintaing new variable in ALL
    skb, even if no tunneling is used.
 
-   Current solution: t->recursion lock breaks dead loops. It looks
-   like dev->tbusy flag, but I preferred new variable, because
-   the semantics is different. One day, when hard_start_xmit
-   will be multithreaded we will have to use skb->encapsulation.
+   Current solution: HARD_TX_LOCK lock breaks dead loops.
 
 
 
@@ -128,7 +125,7 @@ static int ipgre_tunnel_bind_dev(struct net_device *dev);
 
 #define HASH_SIZE  16
 
-static int ipgre_net_id;
+static int ipgre_net_id __read_mostly;
 struct ipgre_net {
 	struct ip_tunnel *tunnels[4][HASH_SIZE];
 
@@ -159,8 +156,13 @@ struct ipgre_net {
 #define tunnels_r	tunnels[2]
 #define tunnels_l	tunnels[1]
 #define tunnels_wc	tunnels[0]
+/*
+ * Locking : hash tables are protected by RCU and a spinlock
+ */
+static DEFINE_SPINLOCK(ipgre_lock);
 
-static DEFINE_RWLOCK(ipgre_lock);
+#define for_each_ip_tunnel_rcu(start) \
+	for (t = rcu_dereference(start); t; t = rcu_dereference(t->next))
 
 /* Given src, dst and key, find appropriate for input tunnel. */
 
@@ -172,13 +174,13 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev,
 	int link = dev->ifindex;
 	unsigned h0 = HASH(remote);
 	unsigned h1 = HASH(key);
-	struct ip_tunnel *t, *sel[4] = { NULL, NULL, NULL, NULL };
+	struct ip_tunnel *t, *cand = NULL;
 	struct ipgre_net *ign = net_generic(net, ipgre_net_id);
 	int dev_type = (gre_proto == htons(ETH_P_TEB)) ?
 		       ARPHRD_ETHER : ARPHRD_IPGRE;
-	int idx;
+	int score, cand_score = 4;
 
-	for (t = ign->tunnels_r_l[h0^h1]; t; t = t->next) {
+	for_each_ip_tunnel_rcu(ign->tunnels_r_l[h0 ^ h1]) {
 		if (local != t->parms.iph.saddr ||
 		    remote != t->parms.iph.daddr ||
 		    key != t->parms.i_key ||
@@ -189,18 +191,21 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev,
 		    t->dev->type != dev_type)
 			continue;
 
-		idx = 0;
+		score = 0;
 		if (t->parms.link != link)
-			idx |= 1;
+			score |= 1;
 		if (t->dev->type != dev_type)
-			idx |= 2;
-		if (idx == 0)
+			score |= 2;
+		if (score == 0)
 			return t;
-		if (sel[idx] == NULL)
-			sel[idx] = t;
+
+		if (score < cand_score) {
+			cand = t;
+			cand_score = score;
+		}
 	}
 
-	for (t = ign->tunnels_r[h0^h1]; t; t = t->next) {
+	for_each_ip_tunnel_rcu(ign->tunnels_r[h0 ^ h1]) {
 		if (remote != t->parms.iph.daddr ||
 		    key != t->parms.i_key ||
 		    !(t->dev->flags & IFF_UP))
@@ -210,18 +215,21 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev,
 		    t->dev->type != dev_type)
 			continue;
 
-		idx = 0;
+		score = 0;
 		if (t->parms.link != link)
-			idx |= 1;
+			score |= 1;
 		if (t->dev->type != dev_type)
-			idx |= 2;
-		if (idx == 0)
+			score |= 2;
+		if (score == 0)
 			return t;
-		if (sel[idx] == NULL)
-			sel[idx] = t;
+
+		if (score < cand_score) {
+			cand = t;
+			cand_score = score;
+		}
 	}
 
-	for (t = ign->tunnels_l[h1]; t; t = t->next) {
+	for_each_ip_tunnel_rcu(ign->tunnels_l[h1]) {
 		if ((local != t->parms.iph.saddr &&
 		     (local != t->parms.iph.daddr ||
 		      !ipv4_is_multicast(local))) ||
@@ -233,18 +241,21 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev,
 		    t->dev->type != dev_type)
 			continue;
 
-		idx = 0;
+		score = 0;
 		if (t->parms.link != link)
-			idx |= 1;
+			score |= 1;
 		if (t->dev->type != dev_type)
-			idx |= 2;
-		if (idx == 0)
+			score |= 2;
+		if (score == 0)
 			return t;
-		if (sel[idx] == NULL)
-			sel[idx] = t;
+
+		if (score < cand_score) {
+			cand = t;
+			cand_score = score;
+		}
 	}
 
-	for (t = ign->tunnels_wc[h1]; t; t = t->next) {
+	for_each_ip_tunnel_rcu(ign->tunnels_wc[h1]) {
 		if (t->parms.i_key != key ||
 		    !(t->dev->flags & IFF_UP))
 			continue;
@@ -253,23 +264,26 @@ static struct ip_tunnel * ipgre_tunnel_lookup(struct net_device *dev,
 		    t->dev->type != dev_type)
 			continue;
 
-		idx = 0;
+		score = 0;
 		if (t->parms.link != link)
-			idx |= 1;
+			score |= 1;
 		if (t->dev->type != dev_type)
-			idx |= 2;
-		if (idx == 0)
+			score |= 2;
+		if (score == 0)
 			return t;
-		if (sel[idx] == NULL)
-			sel[idx] = t;
+
+		if (score < cand_score) {
+			cand = t;
+			cand_score = score;
+		}
 	}
 
-	for (idx = 1; idx < ARRAY_SIZE(sel); idx++)
-		if (sel[idx] != NULL)
-			return sel[idx];
+	if (cand != NULL)
+		return cand;
 
-	if (ign->fb_tunnel_dev->flags & IFF_UP)
-		return netdev_priv(ign->fb_tunnel_dev);
+	dev = ign->fb_tunnel_dev;
+	if (dev->flags & IFF_UP)
+		return netdev_priv(dev);
 
 	return NULL;
 }
@@ -303,10 +317,10 @@ static void ipgre_tunnel_link(struct ipgre_net *ign, struct ip_tunnel *t)
 {
 	struct ip_tunnel **tp = ipgre_bucket(ign, t);
 
+	spin_lock_bh(&ipgre_lock);
 	t->next = *tp;
-	write_lock_bh(&ipgre_lock);
-	*tp = t;
-	write_unlock_bh(&ipgre_lock);
+	rcu_assign_pointer(*tp, t);
+	spin_unlock_bh(&ipgre_lock);
 }
 
 static void ipgre_tunnel_unlink(struct ipgre_net *ign, struct ip_tunnel *t)
@@ -315,9 +329,9 @@ static void ipgre_tunnel_unlink(struct ipgre_net *ign, struct ip_tunnel *t)
 
 	for (tp = ipgre_bucket(ign, t); *tp; tp = &(*tp)->next) {
 		if (t == *tp) {
-			write_lock_bh(&ipgre_lock);
+			spin_lock_bh(&ipgre_lock);
 			*tp = t->next;
-			write_unlock_bh(&ipgre_lock);
+			spin_unlock_bh(&ipgre_lock);
 			break;
 		}
 	}
@@ -468,7 +482,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
 		break;
 	}
 
-	read_lock(&ipgre_lock);
+	rcu_read_lock();
 	t = ipgre_tunnel_lookup(skb->dev, iph->daddr, iph->saddr,
 				flags & GRE_KEY ?
 				*(((__be32 *)p) + (grehlen / 4) - 1) : 0,
@@ -480,13 +494,13 @@ static void ipgre_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(&ipgre_lock);
+	rcu_read_unlock();
 	return;
 }
 
@@ -565,7 +579,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 
 	gre_proto = *(__be16 *)(h + 2);
 
-	read_lock(&ipgre_lock);
+	rcu_read_lock();
 	if ((tunnel = ipgre_tunnel_lookup(skb->dev,
 					  iph->saddr, iph->daddr, key,
 					  gre_proto))) {
@@ -591,7 +605,7 @@ static int ipgre_rcv(struct sk_buff *skb)
 #ifdef CONFIG_NET_IPGRE_BROADCAST
 		if (ipv4_is_multicast(iph->daddr)) {
 			/* Looped back packet, drop it! */
-			if (skb->rtable->fl.iif == 0)
+			if (skb_rtable(skb)->fl.iif == 0)
 				goto drop;
 			stats->multicast++;
 			skb->pkt_type = PACKET_BROADCAST;
@@ -632,30 +646,30 @@ static int ipgre_rcv(struct sk_buff *skb)
 		stats->rx_packets++;
 		stats->rx_bytes += len;
 		skb->dev = tunnel->dev;
-		dst_release(skb->dst);
-		skb->dst = NULL;
+		skb_dst_drop(skb);
 		nf_reset(skb);
 
 		skb_reset_network_header(skb);
 		ipgre_ecn_decapsulate(iph, skb);
 
 		netif_rx(skb);
-		read_unlock(&ipgre_lock);
+		rcu_read_unlock();
 		return(0);
 	}
 	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
 
 drop:
-	read_unlock(&ipgre_lock);
+	rcu_read_unlock();
 drop_nolock:
 	kfree_skb(skb);
 	return(0);
 }
 
-static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
+static netdev_tx_t ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
-	struct net_device_stats *stats = &tunnel->dev->stats;
+	struct net_device_stats *stats = &dev->stats;
+	struct netdev_queue *txq = netdev_get_tx_queue(dev, 0);
 	struct iphdr  *old_iph = ip_hdr(skb);
 	struct iphdr  *tiph;
 	u8     tos;
@@ -668,11 +682,6 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	__be32 dst;
 	int    mtu;
 
-	if (tunnel->recursion++) {
-		stats->collisions++;
-		goto tx_error;
-	}
-
 	if (dev->type == ARPHRD_ETHER)
 		IPCB(skb)->flags = 0;
 
@@ -687,13 +696,13 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	if ((dst = tiph->daddr) == 0) {
 		/* NBMA tunnel */
 
-		if (skb->dst == NULL) {
+		if (skb_dst(skb) == NULL) {
 			stats->tx_fifo_errors++;
 			goto tx_error;
 		}
 
 		if (skb->protocol == htons(ETH_P_IP)) {
-			rt = skb->rtable;
+			rt = skb_rtable(skb);
 			if ((dst = rt->rt_gateway) == 0)
 				goto tx_error_icmp;
 		}
@@ -701,7 +710,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 		else if (skb->protocol == htons(ETH_P_IPV6)) {
 			struct in6_addr *addr6;
 			int addr_type;
-			struct neighbour *neigh = skb->dst->neighbour;
+			struct neighbour *neigh = skb_dst(skb)->neighbour;
 
 			if (neigh == NULL)
 				goto tx_error;
@@ -725,10 +734,10 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	tos = tiph->tos;
-	if (tos&1) {
+	if (tos == 1) {
+		tos = 0;
 		if (skb->protocol == htons(ETH_P_IP))
 			tos = old_iph->tos;
-		tos &= ~1;
 	}
 
 	{
@@ -755,10 +764,10 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (df)
 		mtu = dst_mtu(&rt->u.dst) - dev->hard_header_len - tunnel->hlen;
 	else
-		mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu;
+		mtu = skb_dst(skb) ? dst_mtu(skb_dst(skb)) : dev->mtu;
 
-	if (skb->dst)
-		skb->dst->ops->update_pmtu(skb->dst, mtu);
+	if (skb_dst(skb))
+		skb_dst(skb)->ops->update_pmtu(skb_dst(skb), mtu);
 
 	if (skb->protocol == htons(ETH_P_IP)) {
 		df |= (old_iph->frag_off&htons(IP_DF));
@@ -772,14 +781,14 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 #ifdef CONFIG_IPV6
 	else if (skb->protocol == htons(ETH_P_IPV6)) {
-		struct rt6_info *rt6 = (struct rt6_info *)skb->dst;
+		struct rt6_info *rt6 = (struct rt6_info *)skb_dst(skb);
 
-		if (rt6 && mtu < dst_mtu(skb->dst) && mtu >= IPV6_MIN_MTU) {
+		if (rt6 && mtu < dst_mtu(skb_dst(skb)) && mtu >= IPV6_MIN_MTU) {
 			if ((tunnel->parms.iph.daddr &&
 			     !ipv4_is_multicast(tunnel->parms.iph.daddr)) ||
 			    rt6->rt6i_dst.plen == 128) {
 				rt6->rt6i_flags |= RTF_MODIFIED;
-				skb->dst->metrics[RTAX_MTU-1] = mtu;
+				skb_dst(skb)->metrics[RTAX_MTU-1] = mtu;
 			}
 		}
 
@@ -792,7 +801,8 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 #endif
 
 	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);
@@ -807,10 +817,9 @@ static int ipgre_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);
@@ -825,8 +834,8 @@ static int ipgre_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.
@@ -877,8 +886,7 @@ static int ipgre_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);
@@ -886,8 +894,7 @@ tx_error_icmp:
 tx_error:
 	stats->tx_errors++;
 	dev_kfree_skb(skb);
-	tunnel->recursion--;
-	return 0;
+	return NETDEV_TX_OK;
 }
 
 static int ipgre_tunnel_bind_dev(struct net_device *dev)
@@ -940,7 +947,7 @@ static int ipgre_tunnel_bind_dev(struct net_device *dev)
 			addend += 4;
 	}
 	dev->needed_headroom = addend + hlen;
-	mtu -= dev->hard_header_len - addend;
+	mtu -= dev->hard_header_len + addend;
 
 	if (mtu < 68)
 		mtu = 68;
@@ -1226,6 +1233,7 @@ static void ipgre_tunnel_setup(struct net_device *dev)
 	dev->iflink		= 0;
 	dev->addr_len		= 4;
 	dev->features		|= NETIF_F_NETNS_LOCAL;
+	dev->priv_flags		&= ~IFF_XMIT_DST_RELEASE;
 }
 
 static int ipgre_tunnel_init(struct net_device *dev)
@@ -1276,39 +1284,33 @@ static void ipgre_fb_tunnel_init(struct net_device *dev)
 }
 
 
-static struct net_protocol ipgre_protocol = {
+static const struct net_protocol ipgre_protocol = {
 	.handler	=	ipgre_rcv,
 	.err_handler	=	ipgre_err,
 	.netns_ok	=	1,
 };
 
-static void ipgre_destroy_tunnels(struct ipgre_net *ign)
+static void ipgre_destroy_tunnels(struct ipgre_net *ign, struct list_head *head)
 {
 	int prio;
 
 	for (prio = 0; prio < 4; prio++) {
 		int h;
 		for (h = 0; h < HASH_SIZE; h++) {
-			struct ip_tunnel *t;
-			while ((t = ign->tunnels[prio][h]) != NULL)
-				unregister_netdevice(t->dev);
+			struct ip_tunnel *t = ign->tunnels[prio][h];
+
+			while (t != NULL) {
+				unregister_netdevice_queue(t->dev, head);
+				t = t->next;
+			}
 		}
 	}
 }
 
-static int ipgre_init_net(struct net *net)
+static int __net_init ipgre_init_net(struct net *net)
 {
+	struct ipgre_net *ign = net_generic(net, ipgre_net_id);
 	int err;
-	struct ipgre_net *ign;
-
-	err = -ENOMEM;
-	ign = kzalloc(sizeof(struct ipgre_net), GFP_KERNEL);
-	if (ign == NULL)
-		goto err_alloc;
-
-	err = net_assign_generic(net, ipgre_net_id, ign);
-	if (err < 0)
-		goto err_assign;
 
 	ign->fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel), "gre0",
 					   ipgre_tunnel_setup);
@@ -1329,27 +1331,26 @@ static int ipgre_init_net(struct net *net)
 err_reg_dev:
 	free_netdev(ign->fb_tunnel_dev);
 err_alloc_dev:
-	/* nothing */
-err_assign:
-	kfree(ign);
-err_alloc:
 	return err;
 }
 
-static void ipgre_exit_net(struct net *net)
+static void __net_exit ipgre_exit_net(struct net *net)
 {
 	struct ipgre_net *ign;
+	LIST_HEAD(list);
 
 	ign = net_generic(net, ipgre_net_id);
 	rtnl_lock();
-	ipgre_destroy_tunnels(ign);
+	ipgre_destroy_tunnels(ign, &list);
+	unregister_netdevice_many(&list);
 	rtnl_unlock();
-	kfree(ign);
 }
 
 static struct pernet_operations ipgre_net_ops = {
 	.init = ipgre_init_net,
 	.exit = ipgre_exit_net,
+	.id   = &ipgre_net_id,
+	.size = sizeof(struct ipgre_net),
 };
 
 static int ipgre_tunnel_validate(struct nlattr *tb[], struct nlattr *data[])
@@ -1463,14 +1464,14 @@ static void ipgre_tap_setup(struct net_device *dev)
 
 	ether_setup(dev);
 
-	dev->netdev_ops		= &ipgre_netdev_ops;
+	dev->netdev_ops		= &ipgre_tap_netdev_ops;
 	dev->destructor 	= free_netdev;
 
 	dev->iflink		= 0;
 	dev->features		|= NETIF_F_NETNS_LOCAL;
 }
 
-static int ipgre_newlink(struct net_device *dev, struct nlattr *tb[],
+static int ipgre_newlink(struct net *src_net, struct net_device *dev, struct nlattr *tb[],
 			 struct nlattr *data[])
 {
 	struct ip_tunnel *nt;
@@ -1524,25 +1525,29 @@ static int ipgre_changelink(struct net_device *dev, struct nlattr *tb[],
 		if (t->dev != dev)
 			return -EEXIST;
 	} else {
-		unsigned nflags = 0;
-
 		t = nt;
 
-		if (ipv4_is_multicast(p.iph.daddr))
-			nflags = IFF_BROADCAST;
-		else if (p.iph.daddr)
-			nflags = IFF_POINTOPOINT;
+		if (dev->type != ARPHRD_ETHER) {
+			unsigned nflags = 0;
 
-		if ((dev->flags ^ nflags) &
-		    (IFF_POINTOPOINT | IFF_BROADCAST))
-			return -EINVAL;
+			if (ipv4_is_multicast(p.iph.daddr))
+				nflags = IFF_BROADCAST;
+			else if (p.iph.daddr)
+				nflags = IFF_POINTOPOINT;
+
+			if ((dev->flags ^ nflags) &
+			    (IFF_POINTOPOINT | IFF_BROADCAST))
+				return -EINVAL;
+		}
 
 		ipgre_tunnel_unlink(ign, t);
 		t->parms.iph.saddr = p.iph.saddr;
 		t->parms.iph.daddr = p.iph.daddr;
 		t->parms.i_key = p.i_key;
-		memcpy(dev->dev_addr, &p.iph.saddr, 4);
-		memcpy(dev->broadcast, &p.iph.daddr, 4);
+		if (dev->type != ARPHRD_ETHER) {
+			memcpy(dev->dev_addr, &p.iph.saddr, 4);
+			memcpy(dev->broadcast, &p.iph.daddr, 4);
+		}
 		ipgre_tunnel_link(ign, t);
 		netdev_state_change(dev);
 	}
@@ -1665,7 +1670,7 @@ static int __init ipgre_init(void)
 		return -EAGAIN;
 	}
 
-	err = register_pernet_gen_device(&ipgre_net_id, &ipgre_net_ops);
+	err = register_pernet_device(&ipgre_net_ops);
 	if (err < 0)
 		goto gen_device_failed;
 
@@ -1683,7 +1688,7 @@ out:
 tap_ops_failed:
 	rtnl_link_unregister(&ipgre_link_ops);
 rtnl_link_failed:
-	unregister_pernet_gen_device(ipgre_net_id, &ipgre_net_ops);
+	unregister_pernet_device(&ipgre_net_ops);
 gen_device_failed:
 	inet_del_protocol(&ipgre_protocol, IPPROTO_GRE);
 	goto out;
@@ -1693,7 +1698,7 @@ static void __exit ipgre_fini(void)
 {
 	rtnl_link_unregister(&ipgre_tap_ops);
 	rtnl_link_unregister(&ipgre_link_ops);
-	unregister_pernet_gen_device(ipgre_net_id, &ipgre_net_ops);
+	unregister_pernet_device(&ipgre_net_ops);
 	if (inet_del_protocol(&ipgre_protocol, IPPROTO_GRE) < 0)
 		printk(KERN_INFO "ipgre close: can't remove protocol\n");
 }