[deliverable/linux.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_zones.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>
#include <net/netfilter/nf_log.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_ACCEPT;

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

	*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_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) {
		nf_log_packet(NFPROTO_IPV4, hooknum, skb, in, out, NULL,
			      "nf_ct_%s: dropping packet", helper->name);
		return ret;
	}

	/* adjust seqs for loopback traffic only in outgoing direction */
	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
	    !nf_is_loopback_packet(skb)) {
		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[] = {
	{
		.procname	= "ip_conntrack_max",
		.data		= &nf_conntrack_max,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "ip_conntrack_count",
		.data		= &init_net.ct.count,
		.maxlen		= sizeof(int),
		.mode		= 0444,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "ip_conntrack_buckets",
		.data		= &init_net.ct.htable_size,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0444,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "ip_conntrack_checksum",
		.data		= &init_net.ct.sysctl_checksum,
		.maxlen		= sizeof(int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec,
	},
	{
		.procname	= "ip_conntrack_log_invalid",
		.data		= &init_net.ct.sysctl_log_invalid,
		.maxlen		= sizeof(unsigned int),
		.mode		= 0644,
		.proc_handler	= proc_dointvec_minmax,
		.extra1		= &log_invalid_proto_min,
		.extra2		= &log_invalid_proto_max,
	},
	{ }
};
#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->inet_rcv_saddr;
	tuple.src.u.tcp.port = inet->inet_sport;
	tuple.dst.u3.ip = inet->inet_daddr;
	tuple.dst.u.tcp.port = inet->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), NF_CT_DEFAULT_ZONE, &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) {
		pr_err("nf_conntrack_ipv4: can't register tcp.\n");
		goto cleanup_sockopt;
	}

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

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

	ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4);
	if (ret < 0) {
		pr_err("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) {
		pr_err("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	24	#include <net/netfilter/nf_conntrack_l3proto.h>
5d0aa2cc	25	#include <net/netfilter/nf_conntrack_zones.h>
9fb9cbb1 YK	26	#include <net/netfilter/nf_conntrack_core.h>
9fb9cbb1 YK	27	#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
dd13b010	28	#include <net/netfilter/nf_nat_helper.h>
73e4022f	29	#include <net/netfilter/ipv4/nf_defrag_ipv4.h>
74f7a655	30	#include <net/netfilter/nf_log.h>
dd13b010 PM	31
	32	int (nf_nat_seq_adjust_hook)(struct sk_buff skb,
	33	struct nf_conn *ct,
	34	enum ip_conntrack_info ctinfo);
	35	EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook);
9fb9cbb1	36
8ce8439a JE	37	static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
8ce8439a JE	38	struct nf_conntrack_tuple *tuple)
9fb9cbb1	39	{
32948588 JE	40	const __be32 *ap;
32948588 JE	41	__be32 _addrs[2];
9fb9cbb1 YK	42	ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr),
	43	sizeof(u_int32_t) * 2, _addrs);
	44	if (ap == NULL)
8ce8439a	45	return false;
9fb9cbb1 YK	46
	47	tuple->src.u3.ip = ap[0];
	48	tuple->dst.u3.ip = ap[1];
	49
8ce8439a	50	return true;
9fb9cbb1 YK	51	}
9fb9cbb1 YK	52
8ce8439a JE	53	static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
8ce8439a JE	54	const struct nf_conntrack_tuple *orig)
9fb9cbb1 YK	55	{
	56	tuple->src.u3.ip = orig->dst.u3.ip;
	57	tuple->dst.u3.ip = orig->src.u3.ip;
	58
8ce8439a	59	return true;
9fb9cbb1 YK	60	}
	61
	62	static int ipv4_print_tuple(struct seq_file *s,
	63	const struct nf_conntrack_tuple *tuple)
	64	{
cffee385 HH	65	return seq_printf(s, "src=%pI4 dst=%pI4 ",
cffee385 HH	66	&tuple->src.u3.ip, &tuple->dst.u3.ip);
9fb9cbb1 YK	67	}
9fb9cbb1 YK	68
ffc30690 YK	69	static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
ffc30690 YK	70	unsigned int dataoff, u_int8_t protonum)
9fb9cbb1	71	{
32948588 JE	72	const struct iphdr *iph;
32948588 JE	73	struct iphdr _iph;
ffc30690 YK	74
	75	iph = skb_header_pointer(skb, nhoff, sizeof(_iph), &_iph);
	76	if (iph == NULL)
8430eac2	77	return -NF_ACCEPT;
ffc30690	78
0fb96701 PM	79	/* Conntrack defragments packets, we might still see fragments
	80	* inside ICMP packets though. */
	81	if (iph->frag_off & htons(IP_OFFSET))
8430eac2	82	return -NF_ACCEPT;
9fb9cbb1	83
ffc30690 YK	84	*dataoff = nhoff + (iph->ihl << 2);
ffc30690 YK	85	*protonum = iph->protocol;
9fb9cbb1 YK	86
	87	return NF_ACCEPT;
	88	}
	89
9fb9cbb1	90	static unsigned int ipv4_confirm(unsigned int hooknum,
3db05fea	91	struct sk_buff *skb,
9fb9cbb1 YK	92	const struct net_device *in,
	93	const struct net_device *out,
	94	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	95	{
	96	struct nf_conn *ct;
	97	enum ip_conntrack_info ctinfo;
32948588 JE	98	const struct nf_conn_help *help;
32948588 JE	99	const struct nf_conntrack_helper *helper;
dd13b010	100	unsigned int ret;
9fb9cbb1 YK	101
9fb9cbb1 YK	102	/* This is where we call the helper: as the packet goes out. */
3db05fea	103	ct = nf_ct_get(skb, &ctinfo);
fb048833	104	if (!ct \|\| ctinfo == IP_CT_RELATED_REPLY)
dd13b010	105	goto out;
dc808fe2 HW	106
dc808fe2 HW	107	help = nfct_help(ct);
3c158f7f	108	if (!help)
dd13b010 PM	109	goto out;
dd13b010 PM	110
3c158f7f PM	111	/* rcu_read_lock()ed by nf_hook_slow */
	112	helper = rcu_dereference(help->helper);
	113	if (!helper)
dd13b010 PM	114	goto out;
	115
	116	ret = helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
	117	ct, ctinfo);
74f7a655 PM	118	if (ret != NF_ACCEPT) {
	119	nf_log_packet(NFPROTO_IPV4, hooknum, skb, in, out, NULL,
	120	"nf_ct_%s: dropping packet", helper->name);
dd13b010	121	return ret;
74f7a655	122	}
dd13b010	123
42c1edd3 JA	124	/* adjust seqs for loopback traffic only in outgoing direction */
	125	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
	126	!nf_is_loopback_packet(skb)) {
dd13b010 PM	127	typeof(nf_nat_seq_adjust_hook) seq_adjust;
	128
	129	seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
1db7a748 PNA	130	if (!seq_adjust \|\| !seq_adjust(skb, ct, ctinfo)) {
1db7a748 PNA	131	NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
dd13b010	132	return NF_DROP;
1db7a748	133	}
dd13b010 PM	134	}
	135	out:
	136	/* We've seen it coming out the other side: confirm it */
	137	return nf_conntrack_confirm(skb);
9fb9cbb1 YK	138	}
9fb9cbb1 YK	139
9fb9cbb1	140	static unsigned int ipv4_conntrack_in(unsigned int hooknum,
3db05fea	141	struct sk_buff *skb,
9fb9cbb1 YK	142	const struct net_device *in,
	143	const struct net_device *out,
	144	int (okfn)(struct sk_buff ))
	145	{
a702a65f	146	return nf_conntrack_in(dev_net(in), PF_INET, hooknum, skb);
9fb9cbb1 YK	147	}
	148
	149	static unsigned int ipv4_conntrack_local(unsigned int hooknum,
3db05fea	150	struct sk_buff *skb,
e905a9ed YH	151	const struct net_device *in,
	152	const struct net_device *out,
	153	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	154	{
9fb9cbb1 YK	155	/* root is playing with raw sockets. */
3db05fea	156	if (skb->len < sizeof(struct iphdr) \|\|
88843104	157	ip_hdrlen(skb) < sizeof(struct iphdr))
9fb9cbb1	158	return NF_ACCEPT;
a702a65f	159	return nf_conntrack_in(dev_net(out), PF_INET, hooknum, skb);
9fb9cbb1 YK	160	}
	161
	162	/* Connection tracking may drop packets, but never alters them, so
	163	make it the first hook. */
1999414a	164	static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
964ddaa1 PM	165	{
	166	.hook = ipv4_conntrack_in,
	167	.owner = THIS_MODULE,
57750a22	168	.pf = NFPROTO_IPV4,
6e23ae2a	169	.hooknum = NF_INET_PRE_ROUTING,
964ddaa1 PM	170	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	171	},
964ddaa1 PM	172	{
	173	.hook = ipv4_conntrack_local,
	174	.owner = THIS_MODULE,
57750a22	175	.pf = NFPROTO_IPV4,
6e23ae2a	176	.hooknum = NF_INET_LOCAL_OUT,
964ddaa1 PM	177	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	178	},
964ddaa1 PM	179	{
	180	.hook = ipv4_confirm,
	181	.owner = THIS_MODULE,
57750a22	182	.pf = NFPROTO_IPV4,
6e23ae2a	183	.hooknum = NF_INET_POST_ROUTING,
964ddaa1 PM	184	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
	185	},
	186	{
	187	.hook = ipv4_confirm,
	188	.owner = THIS_MODULE,
57750a22	189	.pf = NFPROTO_IPV4,
6e23ae2a	190	.hooknum = NF_INET_LOCAL_IN,
964ddaa1 PM	191	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
964ddaa1 PM	192	},
9fb9cbb1 YK	193	};
9fb9cbb1 YK	194
a999e683 PM	195	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	196	static int log_invalid_proto_min = 0;
	197	static int log_invalid_proto_max = 255;
	198
	199	static ctl_table ip_ct_sysctl_table[] = {
	200	{
a999e683 PM	201	.procname = "ip_conntrack_max",
	202	.data = &nf_conntrack_max,
	203	.maxlen = sizeof(int),
	204	.mode = 0644,
6d9f239a	205	.proc_handler = proc_dointvec,
a999e683 PM	206	},
a999e683 PM	207	{
a999e683	208	.procname = "ip_conntrack_count",
49ac8713	209	.data = &init_net.ct.count,
a999e683 PM	210	.maxlen = sizeof(int),
a999e683 PM	211	.mode = 0444,
6d9f239a	212	.proc_handler = proc_dointvec,
a999e683 PM	213	},
a999e683 PM	214	{
a999e683	215	.procname = "ip_conntrack_buckets",
d696c7bd	216	.data = &init_net.ct.htable_size,
a999e683 PM	217	.maxlen = sizeof(unsigned int),
a999e683 PM	218	.mode = 0444,
6d9f239a	219	.proc_handler = proc_dointvec,
a999e683 PM	220	},
a999e683 PM	221	{
a999e683	222	.procname = "ip_conntrack_checksum",
c04d0552	223	.data = &init_net.ct.sysctl_checksum,
a999e683 PM	224	.maxlen = sizeof(int),
a999e683 PM	225	.mode = 0644,
6d9f239a	226	.proc_handler = proc_dointvec,
a999e683 PM	227	},
a999e683 PM	228	{
a999e683	229	.procname = "ip_conntrack_log_invalid",
c2a2c7e0	230	.data = &init_net.ct.sysctl_log_invalid,
a999e683 PM	231	.maxlen = sizeof(unsigned int),
a999e683 PM	232	.mode = 0644,
6d9f239a	233	.proc_handler = proc_dointvec_minmax,
a999e683 PM	234	.extra1 = &log_invalid_proto_min,
	235	.extra2 = &log_invalid_proto_max,
	236	},
f8572d8f	237	{ }
a999e683 PM	238	};
	239	#endif /* CONFIG_SYSCTL && CONFIG_NF_CONNTRACK_PROC_COMPAT */
	240
9fb9cbb1 YK	241	/* Fast function for those who don't want to parse /proc (and I don't
	242	blame them). */
	243	/* Reversing the socket's dst/src point of view gives us the reply
	244	mapping. */
	245	static int
	246	getorigdst(struct sock sk, int optval, void __user user, int *len)
	247	{
32948588 JE	248	const struct inet_sock *inet = inet_sk(sk);
32948588 JE	249	const struct nf_conntrack_tuple_hash *h;
9fb9cbb1	250	struct nf_conntrack_tuple tuple;
e905a9ed	251
443a70d5	252	memset(&tuple, 0, sizeof(tuple));
c720c7e8 ED	253	tuple.src.u3.ip = inet->inet_rcv_saddr;
	254	tuple.src.u.tcp.port = inet->inet_sport;
	255	tuple.dst.u3.ip = inet->inet_daddr;
	256	tuple.dst.u.tcp.port = inet->inet_dport;
9fb9cbb1	257	tuple.src.l3num = PF_INET;
54981279	258	tuple.dst.protonum = sk->sk_protocol;
9fb9cbb1	259
54981279 RL	260	/* We only do TCP and SCTP at the moment: is there a better way? */
	261	if (sk->sk_protocol != IPPROTO_TCP && sk->sk_protocol != IPPROTO_SCTP) {
	262	pr_debug("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n");
9fb9cbb1 YK	263	return -ENOPROTOOPT;
	264	}
	265
	266	if ((unsigned int) *len < sizeof(struct sockaddr_in)) {
0d53778e PM	267	pr_debug("SO_ORIGINAL_DST: len %d not %Zu\n",
0d53778e PM	268	*len, sizeof(struct sockaddr_in));
9fb9cbb1 YK	269	return -EINVAL;
	270	}
	271
5d0aa2cc	272	h = nf_conntrack_find_get(sock_net(sk), NF_CT_DEFAULT_ZONE, &tuple);
9fb9cbb1 YK	273	if (h) {
	274	struct sockaddr_in sin;
	275	struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
	276
	277	sin.sin_family = AF_INET;
	278	sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	279	.tuple.dst.u.tcp.port;
	280	sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	281	.tuple.dst.u3.ip;
6c813c3f	282	memset(sin.sin_zero, 0, sizeof(sin.sin_zero));
9fb9cbb1	283
cffee385 HH	284	pr_debug("SO_ORIGINAL_DST: %pI4 %u\n",
cffee385 HH	285	&sin.sin_addr.s_addr, ntohs(sin.sin_port));
9fb9cbb1 YK	286	nf_ct_put(ct);
	287	if (copy_to_user(user, &sin, sizeof(sin)) != 0)
	288	return -EFAULT;
	289	else
	290	return 0;
	291	}
cffee385 HH	292	pr_debug("SO_ORIGINAL_DST: Can't find %pI4/%u-%pI4/%u.\n",
	293	&tuple.src.u3.ip, ntohs(tuple.src.u.tcp.port),
	294	&tuple.dst.u3.ip, ntohs(tuple.dst.u.tcp.port));
9fb9cbb1 YK	295	return -ENOENT;
	296	}
	297
e281db5c	298	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
c1d10adb PNA	299
	300	#include <linux/netfilter/nfnetlink.h>
	301	#include <linux/netfilter/nfnetlink_conntrack.h>
	302
fdf70832	303	static int ipv4_tuple_to_nlattr(struct sk_buff *skb,
c1d10adb PNA	304	const struct nf_conntrack_tuple *tuple)
c1d10adb PNA	305	{
77236b6e PM	306	NLA_PUT_BE32(skb, CTA_IP_V4_SRC, tuple->src.u3.ip);
77236b6e PM	307	NLA_PUT_BE32(skb, CTA_IP_V4_DST, tuple->dst.u3.ip);
c1d10adb PNA	308	return 0;
c1d10adb PNA	309
df6fb868	310	nla_put_failure:
c1d10adb PNA	311	return -1;
	312	}
	313
f73e924c PM	314	static const struct nla_policy ipv4_nla_policy[CTA_IP_MAX+1] = {
	315	[CTA_IP_V4_SRC] = { .type = NLA_U32 },
	316	[CTA_IP_V4_DST] = { .type = NLA_U32 },
c1d10adb PNA	317	};
c1d10adb PNA	318
fdf70832	319	static int ipv4_nlattr_to_tuple(struct nlattr *tb[],
c1d10adb PNA	320	struct nf_conntrack_tuple *t)
c1d10adb PNA	321	{
df6fb868	322	if (!tb[CTA_IP_V4_SRC] \|\| !tb[CTA_IP_V4_DST])
c1d10adb PNA	323	return -EINVAL;
c1d10adb PNA	324
77236b6e PM	325	t->src.u3.ip = nla_get_be32(tb[CTA_IP_V4_SRC]);
77236b6e PM	326	t->dst.u3.ip = nla_get_be32(tb[CTA_IP_V4_DST]);
c1d10adb PNA	327
	328	return 0;
	329	}
a400c30e HE	330
	331	static int ipv4_nlattr_tuple_size(void)
	332	{
	333	return nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1);
	334	}
c1d10adb PNA	335	#endif
c1d10adb PNA	336
9fb9cbb1 YK	337	static struct nf_sockopt_ops so_getorigdst = {
	338	.pf = PF_INET,
	339	.get_optmin = SO_ORIGINAL_DST,
	340	.get_optmax = SO_ORIGINAL_DST+1,
	341	.get = &getorigdst,
16fcec35	342	.owner = THIS_MODULE,
9fb9cbb1 YK	343	};
9fb9cbb1 YK	344
61075af5	345	struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
9fb9cbb1 YK	346	.l3proto = PF_INET,
	347	.name = "ipv4",
	348	.pkt_to_tuple = ipv4_pkt_to_tuple,
	349	.invert_tuple = ipv4_invert_tuple,
	350	.print_tuple = ipv4_print_tuple,
ffc30690	351	.get_l4proto = ipv4_get_l4proto,
e281db5c	352	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
fdf70832	353	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
a400c30e	354	.nlattr_tuple_size = ipv4_nlattr_tuple_size,
fdf70832	355	.nlattr_to_tuple = ipv4_nlattr_to_tuple,
f73e924c	356	.nla_policy = ipv4_nla_policy,
a999e683 PM	357	#endif
	358	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	359	.ctl_table_path = nf_net_ipv4_netfilter_sysctl_path,
	360	.ctl_table = ip_ct_sysctl_table,
c1d10adb	361	#endif
9fb9cbb1 YK	362	.me = THIS_MODULE,
	363	};
	364
fae718dd PM	365	module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint,
	366	&nf_conntrack_htable_size, 0600);
	367
32292a7f	368	MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET));
d2483dde	369	MODULE_ALIAS("ip_conntrack");
32292a7f PM	370	MODULE_LICENSE("GPL");
	371
	372	static int __init nf_conntrack_l3proto_ipv4_init(void)
9fb9cbb1 YK	373	{
	374	int ret = 0;
	375
32292a7f	376	need_conntrack();
73e4022f	377	nf_defrag_ipv4_enable();
9fb9cbb1 YK	378
	379	ret = nf_register_sockopt(&so_getorigdst);
	380	if (ret < 0) {
	381	printk(KERN_ERR "Unable to register netfilter socket option\n");
32292a7f	382	return ret;
9fb9cbb1 YK	383	}
9fb9cbb1 YK	384
605dcad6	385	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp4);
9fb9cbb1	386	if (ret < 0) {
654d0fbd	387	pr_err("nf_conntrack_ipv4: can't register tcp.\n");
9fb9cbb1 YK	388	goto cleanup_sockopt;
	389	}
	390
605dcad6	391	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp4);
9fb9cbb1	392	if (ret < 0) {
654d0fbd	393	pr_err("nf_conntrack_ipv4: can't register udp.\n");
9fb9cbb1 YK	394	goto cleanup_tcp;
	395	}
	396
605dcad6	397	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmp);
9fb9cbb1	398	if (ret < 0) {
654d0fbd	399	pr_err("nf_conntrack_ipv4: can't register icmp.\n");
9fb9cbb1 YK	400	goto cleanup_udp;
	401	}
	402
	403	ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4);
	404	if (ret < 0) {
654d0fbd	405	pr_err("nf_conntrack_ipv4: can't register ipv4\n");
9fb9cbb1 YK	406	goto cleanup_icmp;
	407	}
	408
964ddaa1 PM	409	ret = nf_register_hooks(ipv4_conntrack_ops,
964ddaa1 PM	410	ARRAY_SIZE(ipv4_conntrack_ops));
9fb9cbb1	411	if (ret < 0) {
654d0fbd	412	pr_err("nf_conntrack_ipv4: can't register hooks.\n");
9fb9cbb1 YK	413	goto cleanup_ipv4;
9fb9cbb1 YK	414	}
e4bd8bce PM	415	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	416	ret = nf_conntrack_ipv4_compat_init();
	417	if (ret < 0)
	418	goto cleanup_hooks;
	419	#endif
9fb9cbb1	420	return ret;
e4bd8bce PM	421	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
e4bd8bce PM	422	cleanup_hooks:
e905a9ed	423	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
e4bd8bce	424	#endif
9fb9cbb1 YK	425	cleanup_ipv4:
	426	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
	427	cleanup_icmp:
605dcad6	428	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
9fb9cbb1	429	cleanup_udp:
605dcad6	430	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
9fb9cbb1	431	cleanup_tcp:
605dcad6	432	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	433	cleanup_sockopt:
9fb9cbb1 YK	434	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	435	return ret;
	436	}
	437
65b4b4e8	438	static void __exit nf_conntrack_l3proto_ipv4_fini(void)
9fb9cbb1	439	{
32292a7f	440	synchronize_net();
e4bd8bce PM	441	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	442	nf_conntrack_ipv4_compat_fini();
	443	#endif
32292a7f PM	444	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
32292a7f PM	445	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
605dcad6 MJ	446	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
	447	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
	448	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
32292a7f	449	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	450	}
9fb9cbb1 YK	451
65b4b4e8 AM	452	module_init(nf_conntrack_l3proto_ipv4_init);
65b4b4e8 AM	453	module_exit(nf_conntrack_l3proto_ipv4_fini);
591e6206 PM	454
	455	void need_ipv4_conntrack(void)
	456	{
	457	return;
	458	}
	459	EXPORT_SYMBOL_GPL(need_ipv4_conntrack);