[net-next-2.6.git] / net / ipv4 / netfilter / nf_conntrack_l3proto_ipv4.c


/* (C) 1999-2001 Paul `Rusty' Russell
 * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */

#include <linux/types.h>
#include <linux/ip.h>
#include <linux/netfilter.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/icmp.h>
#include <linux/sysctl.h>
#include <net/route.h>
#include <net/ip.h>

#include <linux/netfilter_ipv4.h>
#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_helper.h>
#include <net/netfilter/nf_conntrack_l4proto.h>
#include <net/netfilter/nf_conntrack_l3proto.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
#include <net/netfilter/nf_nat_helper.h>
#include <net/netfilter/ipv4/nf_defrag_ipv4.h>

int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb,
			      struct nf_conn *ct,
			      enum ip_conntrack_info ctinfo);
EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook);

static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
			      struct nf_conntrack_tuple *tuple)
{
	const __be32 *ap;
	__be32 _addrs[2];
	ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr),
				sizeof(u_int32_t) * 2, _addrs);
	if (ap == NULL)
		return false;

	tuple->src.u3.ip = ap[0];
	tuple->dst.u3.ip = ap[1];

	return true;
}

static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
			      const struct nf_conntrack_tuple *orig)
{
	tuple->src.u3.ip = orig->dst.u3.ip;
	tuple->dst.u3.ip = orig->src.u3.ip;

	return true;
}

static int ipv4_print_tuple(struct seq_file *s,
			    const struct nf_conntrack_tuple *tuple)
{
	return seq_printf(s, "src=%pI4 dst=%pI4 ",
			  &tuple->src.u3.ip, &tuple->dst.u3.ip);
}

static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
			    unsigned int *dataoff, u_int8_t *protonum)
{
	const struct iphdr *iph;
	struct iphdr _iph;

	iph = skb_header_pointer(skb, nhoff, sizeof(_iph), &_iph);
	if (iph == NULL)
		return -NF_DROP;

	/* Conntrack defragments packets, we might still see fragments
	 * inside ICMP packets though. */
	if (iph->frag_off & htons(IP_OFFSET))
		return -NF_DROP;

	*dataoff = nhoff + (iph->ihl << 2);
	*protonum = iph->protocol;

	return NF_ACCEPT;
}

static unsigned int ipv4_confirm(unsigned int hooknum,
				 struct sk_buff *skb,
				 const struct net_device *in,
				 const struct net_device *out,
				 int (*okfn)(struct sk_buff *))
{
	struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	const struct nf_conn_help *help;
	const struct nf_conntrack_helper *helper;
	unsigned int ret;

	/* This is where we call the helper: as the packet goes out. */
	ct = nf_ct_get(skb, &ctinfo);
	if (!ct || ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY)
		goto out;

	help = nfct_help(ct);
	if (!help)
		goto out;

	/* rcu_read_lock()ed by nf_hook_slow */
	helper = rcu_dereference(help->helper);
	if (!helper)
		goto out;

	ret = helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
			   ct, ctinfo);
	if (ret != NF_ACCEPT)
		return ret;

	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
		typeof(nf_nat_seq_adjust_hook) seq_adjust;

		seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
		if (!seq_adjust || !seq_adjust(skb, ct, ctinfo)) {
			NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
			return NF_DROP;
		}
	}
out:
	/* We've seen it coming out the other side: confirm it */
	return nf_conntrack_confirm(skb);
}

static unsigned int ipv4_conntrack_in(unsigned int hooknum,
				      struct sk_buff *skb,
				      const struct net_device *in,
				      const struct net_device *out,
				      int (*okfn)(struct sk_buff *))
{
	return nf_conntrack_in(dev_net(in), PF_INET, hooknum, skb);
}

static unsigned int ipv4_conntrack_local(unsigned int hooknum,
					 struct sk_buff *skb,
					 const struct net_device *in,
					 const struct net_device *out,
					 int (*okfn)(struct sk_buff *))
{
	/* root is playing with raw sockets. */
	if (skb->len < sizeof(struct iphdr) ||
	    ip_hdrlen(skb) < sizeof(struct iphdr))
		return NF_ACCEPT;
	return nf_conntrack_in(dev_net(out), PF_INET, hooknum, skb);
}

/* Connection tracking may drop packets, but never alters them, so
   make it the first hook. */
static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
	{
		.hook		= ipv4_conntrack_in,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_PRE_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK,
	},
	{
		.hook		= ipv4_conntrack_local,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_LOCAL_OUT,
		.priority	= NF_IP_PRI_CONNTRACK,
	},
	{
		.hook		= ipv4_confirm,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_POST_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
	},
	{
		.hook		= ipv4_confirm,
		.owner		= THIS_MODULE,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_LOCAL_IN,
		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
	},
};

#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
static int log_invalid_proto_min = 0;
static int log_invalid_proto_max = 255;

static ctl_table ip_ct_sysctl_table[] = {
	{
		.ctl_name	= NET_IPV4_NF_CONNTRACK_MAX,
		.procname	= "ip_conntrack_max",
		.data		= &nf_conntrack_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
	{
		.ctl_name	= NET_IPV4_NF_CONNTRACK_COUNT,
		.procname	= "ip_conntrack_count",
		.data		= &init_net.ct.count,
		.maxlen		= sizeof(int),
		.mode		= 0444,
		.proc_handler	= proc_dointvec,
	},
	{
		.ctl_name	= NET_IPV4_NF_CONNTRACK_BUCKETS,
		.procname	= "ip_conntrack_buckets",
		.data		= &nf_conntrack_htable_size,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0444,
		.proc_handler	= proc_dointvec,
	},
	{
		.ctl_name	= NET_IPV4_NF_CONNTRACK_CHECKSUM,
		.procname	= "ip_conntrack_checksum",
		.data		= &init_net.ct.sysctl_checksum,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
	{
		.ctl_name	= NET_IPV4_NF_CONNTRACK_LOG_INVALID,
		.procname	= "ip_conntrack_log_invalid",
		.data		= &init_net.ct.sysctl_log_invalid,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.strategy	= sysctl_intvec,
		.extra1		= &log_invalid_proto_min,
		.extra2		= &log_invalid_proto_max,
	},
	{
		.ctl_name	= 0
	}
};
#endif /* CONFIG_SYSCTL && CONFIG_NF_CONNTRACK_PROC_COMPAT */

/* Fast function for those who don't want to parse /proc (and I don't
   blame them). */
/* Reversing the socket's dst/src point of view gives us the reply
   mapping. */
static int
getorigdst(struct sock *sk, int optval, void __user *user, int *len)
{
	const struct inet_sock *inet = inet_sk(sk);
	const struct nf_conntrack_tuple_hash *h;
	struct nf_conntrack_tuple tuple;

	memset(&tuple, 0, sizeof(tuple));
	tuple.src.u3.ip = inet->rcv_saddr;
	tuple.src.u.tcp.port = inet->sport;
	tuple.dst.u3.ip = inet->daddr;
	tuple.dst.u.tcp.port = inet->dport;
	tuple.src.l3num = PF_INET;
	tuple.dst.protonum = sk->sk_protocol;

	/* We only do TCP and SCTP at the moment: is there a better way? */
	if (sk->sk_protocol != IPPROTO_TCP && sk->sk_protocol != IPPROTO_SCTP) {
		pr_debug("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n");
		return -ENOPROTOOPT;
	}

	if ((unsigned int) *len < sizeof(struct sockaddr_in)) {
		pr_debug("SO_ORIGINAL_DST: len %d not %Zu\n",
			 *len, sizeof(struct sockaddr_in));
		return -EINVAL;
	}

	h = nf_conntrack_find_get(sock_net(sk), &tuple);
	if (h) {
		struct sockaddr_in sin;
		struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);

		sin.sin_family = AF_INET;
		sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL]
			.tuple.dst.u.tcp.port;
		sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL]
			.tuple.dst.u3.ip;
		memset(sin.sin_zero, 0, sizeof(sin.sin_zero));

		pr_debug("SO_ORIGINAL_DST: %pI4 %u\n",
			 &sin.sin_addr.s_addr, ntohs(sin.sin_port));
		nf_ct_put(ct);
		if (copy_to_user(user, &sin, sizeof(sin)) != 0)
			return -EFAULT;
		else
			return 0;
	}
	pr_debug("SO_ORIGINAL_DST: Can't find %pI4/%u-%pI4/%u.\n",
		 &tuple.src.u3.ip, ntohs(tuple.src.u.tcp.port),
		 &tuple.dst.u3.ip, ntohs(tuple.dst.u.tcp.port));
	return -ENOENT;
}

#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)

#include <linux/netfilter/nfnetlink.h>
#include <linux/netfilter/nfnetlink_conntrack.h>

static int ipv4_tuple_to_nlattr(struct sk_buff *skb,
				const struct nf_conntrack_tuple *tuple)
{
	NLA_PUT_BE32(skb, CTA_IP_V4_SRC, tuple->src.u3.ip);
	NLA_PUT_BE32(skb, CTA_IP_V4_DST, tuple->dst.u3.ip);
	return 0;

nla_put_failure:
	return -1;
}

static const struct nla_policy ipv4_nla_policy[CTA_IP_MAX+1] = {
	[CTA_IP_V4_SRC]	= { .type = NLA_U32 },
	[CTA_IP_V4_DST]	= { .type = NLA_U32 },
};

static int ipv4_nlattr_to_tuple(struct nlattr *tb[],
				struct nf_conntrack_tuple *t)
{
	if (!tb[CTA_IP_V4_SRC] || !tb[CTA_IP_V4_DST])
		return -EINVAL;

	t->src.u3.ip = nla_get_be32(tb[CTA_IP_V4_SRC]);
	t->dst.u3.ip = nla_get_be32(tb[CTA_IP_V4_DST]);

	return 0;
}

static int ipv4_nlattr_tuple_size(void)
{
	return nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1);
}
#endif

static struct nf_sockopt_ops so_getorigdst = {
	.pf		= PF_INET,
	.get_optmin	= SO_ORIGINAL_DST,
	.get_optmax	= SO_ORIGINAL_DST+1,
	.get		= &getorigdst,
	.owner		= THIS_MODULE,
};

struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
	.l3proto	 = PF_INET,
	.name		 = "ipv4",
	.pkt_to_tuple	 = ipv4_pkt_to_tuple,
	.invert_tuple	 = ipv4_invert_tuple,
	.print_tuple	 = ipv4_print_tuple,
	.get_l4proto	 = ipv4_get_l4proto,
#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE)
	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
	.nlattr_tuple_size = ipv4_nlattr_tuple_size,
	.nlattr_to_tuple = ipv4_nlattr_to_tuple,
	.nla_policy	 = ipv4_nla_policy,
#endif
#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	.ctl_table_path  = nf_net_ipv4_netfilter_sysctl_path,
	.ctl_table	 = ip_ct_sysctl_table,
#endif
	.me		 = THIS_MODULE,
};

module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint,
		  &nf_conntrack_htable_size, 0600);

MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET));
MODULE_ALIAS("ip_conntrack");
MODULE_LICENSE("GPL");

static int __init nf_conntrack_l3proto_ipv4_init(void)
{
	int ret = 0;

	need_conntrack();
	nf_defrag_ipv4_enable();

	ret = nf_register_sockopt(&so_getorigdst);
	if (ret < 0) {
		printk(KERN_ERR "Unable to register netfilter socket option\n");
		return ret;
	}

	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp4);
	if (ret < 0) {
		printk("nf_conntrack_ipv4: can't register tcp.\n");
		goto cleanup_sockopt;
	}

	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp4);
	if (ret < 0) {
		printk("nf_conntrack_ipv4: can't register udp.\n");
		goto cleanup_tcp;
	}

	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmp);
	if (ret < 0) {
		printk("nf_conntrack_ipv4: can't register icmp.\n");
		goto cleanup_udp;
	}

	ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4);
	if (ret < 0) {
		printk("nf_conntrack_ipv4: can't register ipv4\n");
		goto cleanup_icmp;
	}

	ret = nf_register_hooks(ipv4_conntrack_ops,
				ARRAY_SIZE(ipv4_conntrack_ops));
	if (ret < 0) {
		printk("nf_conntrack_ipv4: can't register hooks.\n");
		goto cleanup_ipv4;
	}
#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	ret = nf_conntrack_ipv4_compat_init();
	if (ret < 0)
		goto cleanup_hooks;
#endif
	return ret;
#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
 cleanup_hooks:
	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
#endif
 cleanup_ipv4:
	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
 cleanup_icmp:
	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
 cleanup_udp:
	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
 cleanup_tcp:
	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
 cleanup_sockopt:
	nf_unregister_sockopt(&so_getorigdst);
	return ret;
}

static void __exit nf_conntrack_l3proto_ipv4_fini(void)
{
	synchronize_net();
#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	nf_conntrack_ipv4_compat_fini();
#endif
	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
	nf_unregister_sockopt(&so_getorigdst);
}

module_init(nf_conntrack_l3proto_ipv4_init);
module_exit(nf_conntrack_l3proto_ipv4_fini);

void need_ipv4_conntrack(void)
{
	return;
}
EXPORT_SYMBOL_GPL(need_ipv4_conntrack);
Commit	Line	Data
73e4022f	1
9fb9cbb1 YK	2	/* (C) 1999-2001 Paul `Rusty' Russell
	3	* (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org>
	4	*
	5	* This program is free software; you can redistribute it and/or modify
	6	* it under the terms of the GNU General Public License version 2 as
	7	* published by the Free Software Foundation.
9fb9cbb1 YK	8	*/
9fb9cbb1 YK	9
9fb9cbb1 YK	10	#include <linux/types.h>
	11	#include <linux/ip.h>
	12	#include <linux/netfilter.h>
	13	#include <linux/module.h>
	14	#include <linux/skbuff.h>
	15	#include <linux/icmp.h>
	16	#include <linux/sysctl.h>
0ae2cfe7	17	#include <net/route.h>
9fb9cbb1 YK	18	#include <net/ip.h>
	19
	20	#include <linux/netfilter_ipv4.h>
	21	#include <net/netfilter/nf_conntrack.h>
	22	#include <net/netfilter/nf_conntrack_helper.h>
605dcad6	23	#include <net/netfilter/nf_conntrack_l4proto.h>
9fb9cbb1 YK	24	#include <net/netfilter/nf_conntrack_l3proto.h>
	25	#include <net/netfilter/nf_conntrack_core.h>
	26	#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
dd13b010	27	#include <net/netfilter/nf_nat_helper.h>
73e4022f	28	#include <net/netfilter/ipv4/nf_defrag_ipv4.h>
dd13b010 PM	29
	30	int (nf_nat_seq_adjust_hook)(struct sk_buff skb,
	31	struct nf_conn *ct,
	32	enum ip_conntrack_info ctinfo);
	33	EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook);
9fb9cbb1	34
8ce8439a JE	35	static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
8ce8439a JE	36	struct nf_conntrack_tuple *tuple)
9fb9cbb1	37	{
32948588 JE	38	const __be32 *ap;
32948588 JE	39	__be32 _addrs[2];
9fb9cbb1 YK	40	ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr),
	41	sizeof(u_int32_t) * 2, _addrs);
	42	if (ap == NULL)
8ce8439a	43	return false;
9fb9cbb1 YK	44
	45	tuple->src.u3.ip = ap[0];
	46	tuple->dst.u3.ip = ap[1];
	47
8ce8439a	48	return true;
9fb9cbb1 YK	49	}
9fb9cbb1 YK	50
8ce8439a JE	51	static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
8ce8439a JE	52	const struct nf_conntrack_tuple *orig)
9fb9cbb1 YK	53	{
	54	tuple->src.u3.ip = orig->dst.u3.ip;
	55	tuple->dst.u3.ip = orig->src.u3.ip;
	56
8ce8439a	57	return true;
9fb9cbb1 YK	58	}
	59
	60	static int ipv4_print_tuple(struct seq_file *s,
	61	const struct nf_conntrack_tuple *tuple)
	62	{
cffee385 HH	63	return seq_printf(s, "src=%pI4 dst=%pI4 ",
cffee385 HH	64	&tuple->src.u3.ip, &tuple->dst.u3.ip);
9fb9cbb1 YK	65	}
9fb9cbb1 YK	66
ffc30690 YK	67	static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
ffc30690 YK	68	unsigned int dataoff, u_int8_t protonum)
9fb9cbb1	69	{
32948588 JE	70	const struct iphdr *iph;
32948588 JE	71	struct iphdr _iph;
ffc30690 YK	72
	73	iph = skb_header_pointer(skb, nhoff, sizeof(_iph), &_iph);
	74	if (iph == NULL)
	75	return -NF_DROP;
	76
0fb96701 PM	77	/* Conntrack defragments packets, we might still see fragments
	78	* inside ICMP packets though. */
	79	if (iph->frag_off & htons(IP_OFFSET))
9fb9cbb1	80	return -NF_DROP;
9fb9cbb1	81
ffc30690 YK	82	*dataoff = nhoff + (iph->ihl << 2);
ffc30690 YK	83	*protonum = iph->protocol;
9fb9cbb1 YK	84
	85	return NF_ACCEPT;
	86	}
	87
9fb9cbb1	88	static unsigned int ipv4_confirm(unsigned int hooknum,
3db05fea	89	struct sk_buff *skb,
9fb9cbb1 YK	90	const struct net_device *in,
	91	const struct net_device *out,
	92	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	93	{
	94	struct nf_conn *ct;
	95	enum ip_conntrack_info ctinfo;
32948588 JE	96	const struct nf_conn_help *help;
32948588 JE	97	const struct nf_conntrack_helper *helper;
dd13b010	98	unsigned int ret;
9fb9cbb1 YK	99
9fb9cbb1 YK	100	/* This is where we call the helper: as the packet goes out. */
3db05fea	101	ct = nf_ct_get(skb, &ctinfo);
6442f1cf	102	if (!ct \|\| ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY)
dd13b010	103	goto out;
dc808fe2 HW	104
dc808fe2 HW	105	help = nfct_help(ct);
3c158f7f	106	if (!help)
dd13b010 PM	107	goto out;
dd13b010 PM	108
3c158f7f PM	109	/* rcu_read_lock()ed by nf_hook_slow */
	110	helper = rcu_dereference(help->helper);
	111	if (!helper)
dd13b010 PM	112	goto out;
	113
	114	ret = helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
	115	ct, ctinfo);
	116	if (ret != NF_ACCEPT)
	117	return ret;
	118
	119	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
	120	typeof(nf_nat_seq_adjust_hook) seq_adjust;
	121
	122	seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
1db7a748 PNA	123	if (!seq_adjust \|\| !seq_adjust(skb, ct, ctinfo)) {
1db7a748 PNA	124	NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
dd13b010	125	return NF_DROP;
1db7a748	126	}
dd13b010 PM	127	}
	128	out:
	129	/* We've seen it coming out the other side: confirm it */
	130	return nf_conntrack_confirm(skb);
9fb9cbb1 YK	131	}
9fb9cbb1 YK	132
9fb9cbb1	133	static unsigned int ipv4_conntrack_in(unsigned int hooknum,
3db05fea	134	struct sk_buff *skb,
9fb9cbb1 YK	135	const struct net_device *in,
	136	const struct net_device *out,
	137	int (okfn)(struct sk_buff ))
	138	{
a702a65f	139	return nf_conntrack_in(dev_net(in), PF_INET, hooknum, skb);
9fb9cbb1 YK	140	}
	141
	142	static unsigned int ipv4_conntrack_local(unsigned int hooknum,
3db05fea	143	struct sk_buff *skb,
e905a9ed YH	144	const struct net_device *in,
	145	const struct net_device *out,
	146	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	147	{
9fb9cbb1 YK	148	/* root is playing with raw sockets. */
3db05fea	149	if (skb->len < sizeof(struct iphdr) \|\|
88843104	150	ip_hdrlen(skb) < sizeof(struct iphdr))
9fb9cbb1	151	return NF_ACCEPT;
a702a65f	152	return nf_conntrack_in(dev_net(out), PF_INET, hooknum, skb);
9fb9cbb1 YK	153	}
	154
	155	/* Connection tracking may drop packets, but never alters them, so
	156	make it the first hook. */
1999414a	157	static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
964ddaa1 PM	158	{
	159	.hook = ipv4_conntrack_in,
	160	.owner = THIS_MODULE,
57750a22	161	.pf = NFPROTO_IPV4,
6e23ae2a	162	.hooknum = NF_INET_PRE_ROUTING,
964ddaa1 PM	163	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	164	},
964ddaa1 PM	165	{
	166	.hook = ipv4_conntrack_local,
	167	.owner = THIS_MODULE,
57750a22	168	.pf = NFPROTO_IPV4,
6e23ae2a	169	.hooknum = NF_INET_LOCAL_OUT,
964ddaa1 PM	170	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	171	},
964ddaa1 PM	172	{
	173	.hook = ipv4_confirm,
	174	.owner = THIS_MODULE,
57750a22	175	.pf = NFPROTO_IPV4,
6e23ae2a	176	.hooknum = NF_INET_POST_ROUTING,
964ddaa1 PM	177	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
	178	},
	179	{
	180	.hook = ipv4_confirm,
	181	.owner = THIS_MODULE,
57750a22	182	.pf = NFPROTO_IPV4,
6e23ae2a	183	.hooknum = NF_INET_LOCAL_IN,
964ddaa1 PM	184	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
964ddaa1 PM	185	},
9fb9cbb1 YK	186	};
9fb9cbb1 YK	187
a999e683 PM	188	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	189	static int log_invalid_proto_min = 0;
	190	static int log_invalid_proto_max = 255;
	191
	192	static ctl_table ip_ct_sysctl_table[] = {
	193	{
	194	.ctl_name = NET_IPV4_NF_CONNTRACK_MAX,
	195	.procname = "ip_conntrack_max",
	196	.data = &nf_conntrack_max,
	197	.maxlen = sizeof(int),
	198	.mode = 0644,
6d9f239a	199	.proc_handler = proc_dointvec,
a999e683 PM	200	},
	201	{
	202	.ctl_name = NET_IPV4_NF_CONNTRACK_COUNT,
	203	.procname = "ip_conntrack_count",
49ac8713	204	.data = &init_net.ct.count,
a999e683 PM	205	.maxlen = sizeof(int),
a999e683 PM	206	.mode = 0444,
6d9f239a	207	.proc_handler = proc_dointvec,
a999e683 PM	208	},
	209	{
	210	.ctl_name = NET_IPV4_NF_CONNTRACK_BUCKETS,
	211	.procname = "ip_conntrack_buckets",
	212	.data = &nf_conntrack_htable_size,
	213	.maxlen = sizeof(unsigned int),
	214	.mode = 0444,
6d9f239a	215	.proc_handler = proc_dointvec,
a999e683 PM	216	},
	217	{
	218	.ctl_name = NET_IPV4_NF_CONNTRACK_CHECKSUM,
	219	.procname = "ip_conntrack_checksum",
c04d0552	220	.data = &init_net.ct.sysctl_checksum,
a999e683 PM	221	.maxlen = sizeof(int),
a999e683 PM	222	.mode = 0644,
6d9f239a	223	.proc_handler = proc_dointvec,
a999e683 PM	224	},
	225	{
	226	.ctl_name = NET_IPV4_NF_CONNTRACK_LOG_INVALID,
	227	.procname = "ip_conntrack_log_invalid",
c2a2c7e0	228	.data = &init_net.ct.sysctl_log_invalid,
a999e683 PM	229	.maxlen = sizeof(unsigned int),
a999e683 PM	230	.mode = 0644,
6d9f239a AD	231	.proc_handler = proc_dointvec_minmax,
6d9f239a AD	232	.strategy = sysctl_intvec,
a999e683 PM	233	.extra1 = &log_invalid_proto_min,
	234	.extra2 = &log_invalid_proto_max,
	235	},
	236	{
	237	.ctl_name = 0
	238	}
	239	};
	240	#endif /* CONFIG_SYSCTL && CONFIG_NF_CONNTRACK_PROC_COMPAT */
	241
9fb9cbb1 YK	242	/* Fast function for those who don't want to parse /proc (and I don't
	243	blame them). */
	244	/* Reversing the socket's dst/src point of view gives us the reply
	245	mapping. */
	246	static int
	247	getorigdst(struct sock sk, int optval, void __user user, int *len)
	248	{
32948588 JE	249	const struct inet_sock *inet = inet_sk(sk);
32948588 JE	250	const struct nf_conntrack_tuple_hash *h;
9fb9cbb1	251	struct nf_conntrack_tuple tuple;
e905a9ed	252
443a70d5	253	memset(&tuple, 0, sizeof(tuple));
9fb9cbb1 YK	254	tuple.src.u3.ip = inet->rcv_saddr;
	255	tuple.src.u.tcp.port = inet->sport;
	256	tuple.dst.u3.ip = inet->daddr;
	257	tuple.dst.u.tcp.port = inet->dport;
	258	tuple.src.l3num = PF_INET;
54981279	259	tuple.dst.protonum = sk->sk_protocol;
9fb9cbb1	260
54981279 RL	261	/* We only do TCP and SCTP at the moment: is there a better way? */
	262	if (sk->sk_protocol != IPPROTO_TCP && sk->sk_protocol != IPPROTO_SCTP) {
	263	pr_debug("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n");
9fb9cbb1 YK	264	return -ENOPROTOOPT;
	265	}
	266
	267	if ((unsigned int) *len < sizeof(struct sockaddr_in)) {
0d53778e PM	268	pr_debug("SO_ORIGINAL_DST: len %d not %Zu\n",
0d53778e PM	269	*len, sizeof(struct sockaddr_in));
9fb9cbb1 YK	270	return -EINVAL;
	271	}
	272
400dad39	273	h = nf_conntrack_find_get(sock_net(sk), &tuple);
9fb9cbb1 YK	274	if (h) {
	275	struct sockaddr_in sin;
	276	struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
	277
	278	sin.sin_family = AF_INET;
	279	sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	280	.tuple.dst.u.tcp.port;
	281	sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	282	.tuple.dst.u3.ip;
6c813c3f	283	memset(sin.sin_zero, 0, sizeof(sin.sin_zero));
9fb9cbb1	284
cffee385 HH	285	pr_debug("SO_ORIGINAL_DST: %pI4 %u\n",
cffee385 HH	286	&sin.sin_addr.s_addr, ntohs(sin.sin_port));
9fb9cbb1 YK	287	nf_ct_put(ct);
	288	if (copy_to_user(user, &sin, sizeof(sin)) != 0)
	289	return -EFAULT;
	290	else
	291	return 0;
	292	}
cffee385 HH	293	pr_debug("SO_ORIGINAL_DST: Can't find %pI4/%u-%pI4/%u.\n",
	294	&tuple.src.u3.ip, ntohs(tuple.src.u.tcp.port),
	295	&tuple.dst.u3.ip, ntohs(tuple.dst.u.tcp.port));
9fb9cbb1 YK	296	return -ENOENT;
	297	}
	298
e281db5c	299	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
c1d10adb PNA	300
	301	#include <linux/netfilter/nfnetlink.h>
	302	#include <linux/netfilter/nfnetlink_conntrack.h>
	303
fdf70832	304	static int ipv4_tuple_to_nlattr(struct sk_buff *skb,
c1d10adb PNA	305	const struct nf_conntrack_tuple *tuple)
c1d10adb PNA	306	{
77236b6e PM	307	NLA_PUT_BE32(skb, CTA_IP_V4_SRC, tuple->src.u3.ip);
77236b6e PM	308	NLA_PUT_BE32(skb, CTA_IP_V4_DST, tuple->dst.u3.ip);
c1d10adb PNA	309	return 0;
c1d10adb PNA	310
df6fb868	311	nla_put_failure:
c1d10adb PNA	312	return -1;
	313	}
	314
f73e924c PM	315	static const struct nla_policy ipv4_nla_policy[CTA_IP_MAX+1] = {
	316	[CTA_IP_V4_SRC] = { .type = NLA_U32 },
	317	[CTA_IP_V4_DST] = { .type = NLA_U32 },
c1d10adb PNA	318	};
c1d10adb PNA	319
fdf70832	320	static int ipv4_nlattr_to_tuple(struct nlattr *tb[],
c1d10adb PNA	321	struct nf_conntrack_tuple *t)
c1d10adb PNA	322	{
df6fb868	323	if (!tb[CTA_IP_V4_SRC] \|\| !tb[CTA_IP_V4_DST])
c1d10adb PNA	324	return -EINVAL;
c1d10adb PNA	325
77236b6e PM	326	t->src.u3.ip = nla_get_be32(tb[CTA_IP_V4_SRC]);
77236b6e PM	327	t->dst.u3.ip = nla_get_be32(tb[CTA_IP_V4_DST]);
c1d10adb PNA	328
	329	return 0;
	330	}
a400c30e HE	331
	332	static int ipv4_nlattr_tuple_size(void)
	333	{
	334	return nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1);
	335	}
c1d10adb PNA	336	#endif
c1d10adb PNA	337
9fb9cbb1 YK	338	static struct nf_sockopt_ops so_getorigdst = {
	339	.pf = PF_INET,
	340	.get_optmin = SO_ORIGINAL_DST,
	341	.get_optmax = SO_ORIGINAL_DST+1,
	342	.get = &getorigdst,
16fcec35	343	.owner = THIS_MODULE,
9fb9cbb1 YK	344	};
9fb9cbb1 YK	345
61075af5	346	struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
9fb9cbb1 YK	347	.l3proto = PF_INET,
	348	.name = "ipv4",
	349	.pkt_to_tuple = ipv4_pkt_to_tuple,
	350	.invert_tuple = ipv4_invert_tuple,
	351	.print_tuple = ipv4_print_tuple,
ffc30690	352	.get_l4proto = ipv4_get_l4proto,
e281db5c	353	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
fdf70832	354	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
a400c30e	355	.nlattr_tuple_size = ipv4_nlattr_tuple_size,
fdf70832	356	.nlattr_to_tuple = ipv4_nlattr_to_tuple,
f73e924c	357	.nla_policy = ipv4_nla_policy,
a999e683 PM	358	#endif
	359	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	360	.ctl_table_path = nf_net_ipv4_netfilter_sysctl_path,
	361	.ctl_table = ip_ct_sysctl_table,
c1d10adb	362	#endif
9fb9cbb1 YK	363	.me = THIS_MODULE,
	364	};
	365
fae718dd PM	366	module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint,
	367	&nf_conntrack_htable_size, 0600);
	368
32292a7f	369	MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET));
d2483dde	370	MODULE_ALIAS("ip_conntrack");
32292a7f PM	371	MODULE_LICENSE("GPL");
	372
	373	static int __init nf_conntrack_l3proto_ipv4_init(void)
9fb9cbb1 YK	374	{
	375	int ret = 0;
	376
32292a7f	377	need_conntrack();
73e4022f	378	nf_defrag_ipv4_enable();
9fb9cbb1 YK	379
	380	ret = nf_register_sockopt(&so_getorigdst);
	381	if (ret < 0) {
	382	printk(KERN_ERR "Unable to register netfilter socket option\n");
32292a7f	383	return ret;
9fb9cbb1 YK	384	}
9fb9cbb1 YK	385
605dcad6	386	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	387	if (ret < 0) {
	388	printk("nf_conntrack_ipv4: can't register tcp.\n");
	389	goto cleanup_sockopt;
	390	}
	391
605dcad6	392	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp4);
9fb9cbb1 YK	393	if (ret < 0) {
	394	printk("nf_conntrack_ipv4: can't register udp.\n");
	395	goto cleanup_tcp;
	396	}
	397
605dcad6	398	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmp);
9fb9cbb1 YK	399	if (ret < 0) {
	400	printk("nf_conntrack_ipv4: can't register icmp.\n");
	401	goto cleanup_udp;
	402	}
	403
	404	ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4);
	405	if (ret < 0) {
	406	printk("nf_conntrack_ipv4: can't register ipv4\n");
	407	goto cleanup_icmp;
	408	}
	409
964ddaa1 PM	410	ret = nf_register_hooks(ipv4_conntrack_ops,
964ddaa1 PM	411	ARRAY_SIZE(ipv4_conntrack_ops));
9fb9cbb1	412	if (ret < 0) {
964ddaa1	413	printk("nf_conntrack_ipv4: can't register hooks.\n");
9fb9cbb1 YK	414	goto cleanup_ipv4;
9fb9cbb1 YK	415	}
e4bd8bce PM	416	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	417	ret = nf_conntrack_ipv4_compat_init();
	418	if (ret < 0)
	419	goto cleanup_hooks;
	420	#endif
9fb9cbb1	421	return ret;
e4bd8bce PM	422	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
e4bd8bce PM	423	cleanup_hooks:
e905a9ed	424	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
e4bd8bce	425	#endif
9fb9cbb1 YK	426	cleanup_ipv4:
	427	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
	428	cleanup_icmp:
605dcad6	429	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
9fb9cbb1	430	cleanup_udp:
605dcad6	431	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
9fb9cbb1	432	cleanup_tcp:
605dcad6	433	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	434	cleanup_sockopt:
9fb9cbb1 YK	435	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	436	return ret;
	437	}
	438
65b4b4e8	439	static void __exit nf_conntrack_l3proto_ipv4_fini(void)
9fb9cbb1	440	{
32292a7f	441	synchronize_net();
e4bd8bce PM	442	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	443	nf_conntrack_ipv4_compat_fini();
	444	#endif
32292a7f PM	445	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
32292a7f PM	446	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
605dcad6 MJ	447	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
	448	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
	449	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
32292a7f	450	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	451	}
9fb9cbb1 YK	452
65b4b4e8 AM	453	module_init(nf_conntrack_l3proto_ipv4_init);
65b4b4e8 AM	454	module_exit(nf_conntrack_l3proto_ipv4_fini);
591e6206 PM	455
	456	void need_ipv4_conntrack(void)
	457	{
	458	return;
	459	}
	460	EXPORT_SYMBOL_GPL(need_ipv4_conntrack);