X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=net%2Fipv4%2Ficmp.c;h=ff9a8e643fccbe1ee08e76f9190b6409bd07cc59;hb=7524d7d6de5d5d3f081de8cf5479819fad339661;hp=272c69e106e9a6a416249fcb77607ac43c16940a;hpb=541010e4b8921cd781ff02ae68028501457045b6;p=linux-2.6

diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 272c69e106..ff9a8e643f 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -92,6 +92,7 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 #include <net/checksum.h>
+#include <net/xfrm.h>
 
 /*
  *	Build xmit assembly blocks
@@ -228,14 +229,16 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES+1];
  *
  *	On SMP we have one ICMP socket per-cpu.
  */
-static DEFINE_PER_CPU(struct socket *, __icmp_socket) = NULL;
-#define icmp_socket	__get_cpu_var(__icmp_socket)
+static struct sock *icmp_sk(struct net *net)
+{
+	return net->ipv4.icmp_sk[smp_processor_id()];
+}
 
-static __inline__ int icmp_xmit_lock(void)
+static inline int icmp_xmit_lock(struct sock *sk)
 {
 	local_bh_disable();
 
-	if (unlikely(!spin_trylock(&icmp_socket->sk->sk_lock.slock))) {
+	if (unlikely(!spin_trylock(&sk->sk_lock.slock))) {
 		/* This can happen if the output path signals a
 		 * dst_link_failure() for an outgoing ICMP packet.
 		 */
@@ -245,9 +248,9 @@ static __inline__ int icmp_xmit_lock(void)
 	return 0;
 }
 
-static void icmp_xmit_unlock(void)
+static inline void icmp_xmit_unlock(struct sock *sk)
 {
-	spin_unlock_bh(&icmp_socket->sk->sk_lock.slock);
+	spin_unlock_bh(&sk->sk_lock.slock);
 }
 
 /*
@@ -274,18 +277,19 @@ static void icmp_xmit_unlock(void)
 #define XRLIM_BURST_FACTOR 6
 int xrlim_allow(struct dst_entry *dst, int timeout)
 {
-	unsigned long now;
+	unsigned long now, token = dst->rate_tokens;
 	int rc = 0;
 
 	now = jiffies;
-	dst->rate_tokens += now - dst->rate_last;
+	token += now - dst->rate_last;
 	dst->rate_last = now;
-	if (dst->rate_tokens > XRLIM_BURST_FACTOR * timeout)
-		dst->rate_tokens = XRLIM_BURST_FACTOR * timeout;
-	if (dst->rate_tokens >= timeout) {
-		dst->rate_tokens -= timeout;
+	if (token > XRLIM_BURST_FACTOR * timeout)
+		token = XRLIM_BURST_FACTOR * timeout;
+	if (token >= timeout) {
+		token -= timeout;
 		rc = 1;
 	}
+	dst->rate_tokens = token;
 	return rc;
 }
 
@@ -344,19 +348,21 @@ static int icmp_glue_bits(void *from, char *to, int offset, int len, int odd,
 static void icmp_push_reply(struct icmp_bxm *icmp_param,
 			    struct ipcm_cookie *ipc, struct rtable *rt)
 {
+	struct sock *sk;
 	struct sk_buff *skb;
 
-	if (ip_append_data(icmp_socket->sk, icmp_glue_bits, icmp_param,
+	sk = icmp_sk(rt->u.dst.dev->nd_net);
+	if (ip_append_data(sk, icmp_glue_bits, icmp_param,
 			   icmp_param->data_len+icmp_param->head_len,
 			   icmp_param->head_len,
 			   ipc, rt, MSG_DONTWAIT) < 0)
-		ip_flush_pending_frames(icmp_socket->sk);
-	else if ((skb = skb_peek(&icmp_socket->sk->sk_write_queue)) != NULL) {
+		ip_flush_pending_frames(sk);
+	else if ((skb = skb_peek(&sk->sk_write_queue)) != NULL) {
 		struct icmphdr *icmph = icmp_hdr(skb);
 		__wsum csum = 0;
 		struct sk_buff *skb1;
 
-		skb_queue_walk(&icmp_socket->sk->sk_write_queue, skb1) {
+		skb_queue_walk(&sk->sk_write_queue, skb1) {
 			csum = csum_add(csum, skb1->csum);
 		}
 		csum = csum_partial_copy_nocheck((void *)&icmp_param->data,
@@ -364,7 +370,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 						 icmp_param->head_len, csum);
 		icmph->checksum = csum_fold(csum);
 		skb->ip_summed = CHECKSUM_NONE;
-		ip_push_pending_frames(icmp_socket->sk);
+		ip_push_pending_frames(sk);
 	}
 }
 
@@ -374,16 +380,17 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
 
 static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 {
-	struct sock *sk = icmp_socket->sk;
-	struct inet_sock *inet = inet_sk(sk);
 	struct ipcm_cookie ipc;
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
+	struct net *net = rt->u.dst.dev->nd_net;
+	struct sock *sk = icmp_sk(net);
+	struct inet_sock *inet = inet_sk(sk);
 	__be32 daddr;
 
 	if (ip_options_echo(&icmp_param->replyopts, skb))
 		return;
 
-	if (icmp_xmit_lock())
+	if (icmp_xmit_lock(sk))
 		return;
 
 	icmp_param->data.icmph.checksum = 0;
@@ -403,7 +410,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 						.tos = RT_TOS(ip_hdr(skb)->tos) } },
 				    .proto = IPPROTO_ICMP };
 		security_skb_classify_flow(skb, &fl);
-		if (ip_route_output_key(&rt, &fl))
+		if (ip_route_output_key(net, &rt, &fl))
 			goto out_unlock;
 	}
 	if (icmpv4_xrlim_allow(rt, icmp_param->data.icmph.type,
@@ -411,7 +418,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 		icmp_push_reply(icmp_param, &ipc, rt);
 	ip_rt_put(rt);
 out_unlock:
-	icmp_xmit_unlock();
+	icmp_xmit_unlock(sk);
 }
 
 
@@ -431,13 +438,17 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 	struct iphdr *iph;
 	int room;
 	struct icmp_bxm icmp_param;
-	struct rtable *rt = (struct rtable *)skb_in->dst;
+	struct rtable *rt = skb_in->rtable;
 	struct ipcm_cookie ipc;
 	__be32 saddr;
 	u8  tos;
+	struct net *net;
+	struct sock *sk;
 
 	if (!rt)
 		goto out;
+	net = rt->u.dst.dev->nd_net;
+	sk = icmp_sk(net);
 
 	/*
 	 *	Find the original header. It is expected to be valid, of course.
@@ -501,7 +512,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 		}
 	}
 
-	if (icmp_xmit_lock())
+	if (icmp_xmit_lock(sk))
 		return;
 
 	/*
@@ -513,7 +524,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 		struct net_device *dev = NULL;
 
 		if (rt->fl.iif && sysctl_icmp_errors_use_inbound_ifaddr)
-			dev = dev_get_by_index(&init_net, rt->fl.iif);
+			dev = dev_get_by_index(net, rt->fl.iif);
 
 		if (dev) {
 			saddr = inet_select_addr(dev, 0, RT_SCOPE_LINK);
@@ -540,8 +551,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 	icmp_param.data.icmph.checksum	 = 0;
 	icmp_param.skb	  = skb_in;
 	icmp_param.offset = skb_network_offset(skb_in);
-	icmp_out_count(icmp_param.data.icmph.type);
-	inet_sk(icmp_socket->sk)->tos = tos;
+	inet_sk(sk)->tos = tos;
 	ipc.addr = iph->saddr;
 	ipc.opt = &icmp_param.replyopts;
 
@@ -564,11 +574,71 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 				}
 			}
 		};
+		int err;
+		struct rtable *rt2;
+
 		security_skb_classify_flow(skb_in, &fl);
-		if (ip_route_output_key(&rt, &fl))
+		if (__ip_route_output_key(net, &rt, &fl))
+			goto out_unlock;
+
+		/* No need to clone since we're just using its address. */
+		rt2 = rt;
+
+		err = xfrm_lookup((struct dst_entry **)&rt, &fl, NULL, 0);
+		switch (err) {
+		case 0:
+			if (rt != rt2)
+				goto route_done;
+			break;
+		case -EPERM:
+			rt = NULL;
+			break;
+		default:
+			goto out_unlock;
+		}
+
+		if (xfrm_decode_session_reverse(skb_in, &fl, AF_INET))
+			goto out_unlock;
+
+		if (inet_addr_type(net, fl.fl4_src) == RTN_LOCAL)
+			err = __ip_route_output_key(net, &rt2, &fl);
+		else {
+			struct flowi fl2 = {};
+			struct dst_entry *odst;
+
+			fl2.fl4_dst = fl.fl4_src;
+			if (ip_route_output_key(net, &rt2, &fl2))
+				goto out_unlock;
+
+			/* Ugh! */
+			odst = skb_in->dst;
+			err = ip_route_input(skb_in, fl.fl4_dst, fl.fl4_src,
+					     RT_TOS(tos), rt2->u.dst.dev);
+
+			dst_release(&rt2->u.dst);
+			rt2 = skb_in->rtable;
+			skb_in->dst = odst;
+		}
+
+		if (err)
+			goto out_unlock;
+
+		err = xfrm_lookup((struct dst_entry **)&rt2, &fl, NULL,
+				  XFRM_LOOKUP_ICMP);
+		if (err == -ENOENT) {
+			if (!rt)
+				goto out_unlock;
+			goto route_done;
+		}
+
+		dst_release(&rt->u.dst);
+		rt = rt2;
+
+		if (err)
 			goto out_unlock;
 	}
 
+route_done:
 	if (!icmpv4_xrlim_allow(rt, type, code))
 		goto ende;
 
@@ -589,7 +659,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info)
 ende:
 	ip_rt_put(rt);
 out_unlock:
-	icmp_xmit_unlock();
+	icmp_xmit_unlock(sk);
 out:;
 }
 
@@ -604,8 +674,10 @@ static void icmp_unreach(struct sk_buff *skb)
 	struct icmphdr *icmph;
 	int hash, protocol;
 	struct net_protocol *ipprot;
-	struct sock *raw_sk;
 	u32 info = 0;
+	struct net *net;
+
+	net = skb->dst->dev->nd_net;
 
 	/*
 	 *	Incomplete header ?
@@ -636,7 +708,7 @@ static void icmp_unreach(struct sk_buff *skb)
 							 "and DF set.\n",
 					       NIPQUAD(iph->daddr));
 			} else {
-				info = ip_rt_frag_needed(iph,
+				info = ip_rt_frag_needed(net, iph,
 						     ntohs(icmph->un.frag.mtu));
 				if (!info)
 					goto out;
@@ -674,7 +746,7 @@ static void icmp_unreach(struct sk_buff *skb)
 	 */
 
 	if (!sysctl_icmp_ignore_bogus_error_responses &&
-	    inet_addr_type(iph->daddr) == RTN_BROADCAST) {
+	    inet_addr_type(net, iph->daddr) == RTN_BROADCAST) {
 		if (net_ratelimit())
 			printk(KERN_WARNING "%u.%u.%u.%u sent an invalid ICMP "
 					    "type %u, code %u "
@@ -698,21 +770,9 @@ static void icmp_unreach(struct sk_buff *skb)
 	/*
 	 *	Deliver ICMP message to raw sockets. Pretty useless feature?
 	 */
+	raw_icmp_error(skb, protocol, info);
 
-	/* Note: See raw.c and net/raw.h, RAWV4_HTABLE_SIZE==MAX_INET_PROTOS */
 	hash = protocol & (MAX_INET_PROTOS - 1);
-	read_lock(&raw_v4_lock);
-	if ((raw_sk = sk_head(&raw_v4_htable[hash])) != NULL) {
-		while ((raw_sk = __raw_v4_lookup(raw_sk, protocol, iph->daddr,
-						 iph->saddr,
-						 skb->dev->ifindex)) != NULL) {
-			raw_err(raw_sk, skb, info);
-			raw_sk = sk_next(raw_sk);
-			iph = (struct iphdr *)skb->data;
-		}
-	}
-	read_unlock(&raw_v4_lock);
-
 	rcu_read_lock();
 	ipprot = rcu_dereference(inet_protos[hash]);
 	if (ipprot && ipprot->err_handler)
@@ -883,7 +943,7 @@ static void icmp_address(struct sk_buff *skb)
 
 static void icmp_address_reply(struct sk_buff *skb)
 {
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
 	struct net_device *dev = skb->dev;
 	struct in_device *in_dev;
 	struct in_ifaddr *ifa;
@@ -928,7 +988,26 @@ static void icmp_discard(struct sk_buff *skb)
 int icmp_rcv(struct sk_buff *skb)
 {
 	struct icmphdr *icmph;
-	struct rtable *rt = (struct rtable *)skb->dst;
+	struct rtable *rt = skb->rtable;
+
+	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) {
+		int nh;
+
+		if (!(skb->sp && skb->sp->xvec[skb->sp->len - 1]->props.flags &
+				 XFRM_STATE_ICMP))
+			goto drop;
+
+		if (!pskb_may_pull(skb, sizeof(*icmph) + sizeof(struct iphdr)))
+			goto drop;
+
+		nh = skb_network_offset(skb);
+		skb_set_network_header(skb, sizeof(*icmph));
+
+		if (!xfrm4_policy_check_reverse(NULL, XFRM_POLICY_IN, skb))
+			goto drop;
+
+		skb_set_network_header(skb, nh);
+	}
 
 	ICMP_INC_STATS_BH(ICMP_MIB_INMSGS);
 
@@ -943,7 +1022,7 @@ int icmp_rcv(struct sk_buff *skb)
 			goto error;
 	}
 
-	if (!pskb_pull(skb, sizeof(struct icmphdr)))
+	if (!pskb_pull(skb, sizeof(*icmph)))
 		goto error;
 
 	icmph = icmp_hdr(skb);
@@ -1067,29 +1146,46 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = {
 	},
 };
 
-void __init icmp_init(struct net_proto_family *ops)
+static void __net_exit icmp_sk_exit(struct net *net)
 {
-	struct inet_sock *inet;
 	int i;
 
-	for_each_possible_cpu(i) {
-		int err;
+	for_each_possible_cpu(i)
+		sk_release_kernel(net->ipv4.icmp_sk[i]);
+	kfree(net->ipv4.icmp_sk);
+	net->ipv4.icmp_sk = NULL;
+}
+
+int __net_init icmp_sk_init(struct net *net)
+{
+	int i, err;
 
-		err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP,
-				       &per_cpu(__icmp_socket, i));
+	net->ipv4.icmp_sk =
+		kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL);
+	if (net->ipv4.icmp_sk == NULL)
+		return -ENOMEM;
 
+	for_each_possible_cpu(i) {
+		struct sock *sk;
+		struct socket *sock;
+		struct inet_sock *inet;
+
+		err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP, &sock);
 		if (err < 0)
-			panic("Failed to create the ICMP control socket.\n");
+			goto fail;
+
+		net->ipv4.icmp_sk[i] = sk = sock->sk;
+		sk_change_net(sk, net);
 
-		per_cpu(__icmp_socket, i)->sk->sk_allocation = GFP_ATOMIC;
+		sk->sk_allocation = GFP_ATOMIC;
 
 		/* Enough space for 2 64K ICMP packets, including
 		 * sk_buff struct overhead.
 		 */
-		per_cpu(__icmp_socket, i)->sk->sk_sndbuf =
+		sk->sk_sndbuf =
 			(2 * ((64 * 1024) + sizeof(struct sk_buff)));
 
-		inet = inet_sk(per_cpu(__icmp_socket, i)->sk);
+		inet = inet_sk(sk);
 		inet->uc_ttl = -1;
 		inet->pmtudisc = IP_PMTUDISC_DONT;
 
@@ -1097,12 +1193,28 @@ void __init icmp_init(struct net_proto_family *ops)
 		 * see it, we do not wish this socket to see incoming
 		 * packets.
 		 */
-		per_cpu(__icmp_socket, i)->sk->sk_prot->unhash(per_cpu(__icmp_socket, i)->sk);
+		sk->sk_prot->unhash(sk);
 	}
+	return 0;
+
+fail:
+	for_each_possible_cpu(i)
+		sk_release_kernel(net->ipv4.icmp_sk[i]);
+	kfree(net->ipv4.icmp_sk);
+	return err;
+}
+
+static struct pernet_operations __net_initdata icmp_sk_ops = {
+       .init = icmp_sk_init,
+       .exit = icmp_sk_exit,
+};
+
+int __init icmp_init(void)
+{
+	return register_pernet_device(&icmp_sk_ops);
 }
 
 EXPORT_SYMBOL(icmp_err_convert);
 EXPORT_SYMBOL(icmp_send);
 EXPORT_SYMBOL(icmp_statistics);
-EXPORT_SYMBOL(icmpmsg_statistics);
 EXPORT_SYMBOL(xrlim_allow);