[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>
#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_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) {
		nf_log_packet(NFPROTO_IPV4, hooknum, skb, in, out, NULL,
			      "nf_ct_%s: dropping packet", helper->name);
		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[] = {
	{
		.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), &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>
74f7a655	29	#include <net/netfilter/nf_log.h>
dd13b010 PM	30
	31	int (nf_nat_seq_adjust_hook)(struct sk_buff skb,
	32	struct nf_conn *ct,
	33	enum ip_conntrack_info ctinfo);
	34	EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook);
9fb9cbb1	35
8ce8439a JE	36	static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
8ce8439a JE	37	struct nf_conntrack_tuple *tuple)
9fb9cbb1	38	{
32948588 JE	39	const __be32 *ap;
32948588 JE	40	__be32 _addrs[2];
9fb9cbb1 YK	41	ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr),
	42	sizeof(u_int32_t) * 2, _addrs);
	43	if (ap == NULL)
8ce8439a	44	return false;
9fb9cbb1 YK	45
	46	tuple->src.u3.ip = ap[0];
	47	tuple->dst.u3.ip = ap[1];
	48
8ce8439a	49	return true;
9fb9cbb1 YK	50	}
9fb9cbb1 YK	51
8ce8439a JE	52	static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
8ce8439a JE	53	const struct nf_conntrack_tuple *orig)
9fb9cbb1 YK	54	{
	55	tuple->src.u3.ip = orig->dst.u3.ip;
	56	tuple->dst.u3.ip = orig->src.u3.ip;
	57
8ce8439a	58	return true;
9fb9cbb1 YK	59	}
	60
	61	static int ipv4_print_tuple(struct seq_file *s,
	62	const struct nf_conntrack_tuple *tuple)
	63	{
cffee385 HH	64	return seq_printf(s, "src=%pI4 dst=%pI4 ",
cffee385 HH	65	&tuple->src.u3.ip, &tuple->dst.u3.ip);
9fb9cbb1 YK	66	}
9fb9cbb1 YK	67
ffc30690 YK	68	static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
ffc30690 YK	69	unsigned int dataoff, u_int8_t protonum)
9fb9cbb1	70	{
32948588 JE	71	const struct iphdr *iph;
32948588 JE	72	struct iphdr _iph;
ffc30690 YK	73
	74	iph = skb_header_pointer(skb, nhoff, sizeof(_iph), &_iph);
	75	if (iph == NULL)
	76	return -NF_DROP;
	77
0fb96701 PM	78	/* Conntrack defragments packets, we might still see fragments
	79	* inside ICMP packets though. */
	80	if (iph->frag_off & htons(IP_OFFSET))
9fb9cbb1	81	return -NF_DROP;
9fb9cbb1	82
ffc30690 YK	83	*dataoff = nhoff + (iph->ihl << 2);
ffc30690 YK	84	*protonum = iph->protocol;
9fb9cbb1 YK	85
	86	return NF_ACCEPT;
	87	}
	88
9fb9cbb1	89	static unsigned int ipv4_confirm(unsigned int hooknum,
3db05fea	90	struct sk_buff *skb,
9fb9cbb1 YK	91	const struct net_device *in,
	92	const struct net_device *out,
	93	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	94	{
	95	struct nf_conn *ct;
	96	enum ip_conntrack_info ctinfo;
32948588 JE	97	const struct nf_conn_help *help;
32948588 JE	98	const struct nf_conntrack_helper *helper;
dd13b010	99	unsigned int ret;
9fb9cbb1 YK	100
9fb9cbb1 YK	101	/* This is where we call the helper: as the packet goes out. */
3db05fea	102	ct = nf_ct_get(skb, &ctinfo);
6442f1cf	103	if (!ct \|\| ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY)
dd13b010	104	goto out;
dc808fe2 HW	105
dc808fe2 HW	106	help = nfct_help(ct);
3c158f7f	107	if (!help)
dd13b010 PM	108	goto out;
dd13b010 PM	109
3c158f7f PM	110	/* rcu_read_lock()ed by nf_hook_slow */
	111	helper = rcu_dereference(help->helper);
	112	if (!helper)
dd13b010 PM	113	goto out;
	114
	115	ret = helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
	116	ct, ctinfo);
74f7a655 PM	117	if (ret != NF_ACCEPT) {
	118	nf_log_packet(NFPROTO_IPV4, hooknum, skb, in, out, NULL,
	119	"nf_ct_%s: dropping packet", helper->name);
dd13b010	120	return ret;
74f7a655	121	}
dd13b010 PM	122
	123	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
	124	typeof(nf_nat_seq_adjust_hook) seq_adjust;
	125
	126	seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook);
1db7a748 PNA	127	if (!seq_adjust \|\| !seq_adjust(skb, ct, ctinfo)) {
1db7a748 PNA	128	NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
dd13b010	129	return NF_DROP;
1db7a748	130	}
dd13b010 PM	131	}
	132	out:
	133	/* We've seen it coming out the other side: confirm it */
	134	return nf_conntrack_confirm(skb);
9fb9cbb1 YK	135	}
9fb9cbb1 YK	136
9fb9cbb1	137	static unsigned int ipv4_conntrack_in(unsigned int hooknum,
3db05fea	138	struct sk_buff *skb,
9fb9cbb1 YK	139	const struct net_device *in,
	140	const struct net_device *out,
	141	int (okfn)(struct sk_buff ))
	142	{
a702a65f	143	return nf_conntrack_in(dev_net(in), PF_INET, hooknum, skb);
9fb9cbb1 YK	144	}
	145
	146	static unsigned int ipv4_conntrack_local(unsigned int hooknum,
3db05fea	147	struct sk_buff *skb,
e905a9ed YH	148	const struct net_device *in,
	149	const struct net_device *out,
	150	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	151	{
9fb9cbb1 YK	152	/* root is playing with raw sockets. */
3db05fea	153	if (skb->len < sizeof(struct iphdr) \|\|
88843104	154	ip_hdrlen(skb) < sizeof(struct iphdr))
9fb9cbb1	155	return NF_ACCEPT;
a702a65f	156	return nf_conntrack_in(dev_net(out), PF_INET, hooknum, skb);
9fb9cbb1 YK	157	}
	158
	159	/* Connection tracking may drop packets, but never alters them, so
	160	make it the first hook. */
1999414a	161	static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
964ddaa1 PM	162	{
	163	.hook = ipv4_conntrack_in,
	164	.owner = THIS_MODULE,
57750a22	165	.pf = NFPROTO_IPV4,
6e23ae2a	166	.hooknum = NF_INET_PRE_ROUTING,
964ddaa1 PM	167	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	168	},
964ddaa1 PM	169	{
	170	.hook = ipv4_conntrack_local,
	171	.owner = THIS_MODULE,
57750a22	172	.pf = NFPROTO_IPV4,
6e23ae2a	173	.hooknum = NF_INET_LOCAL_OUT,
964ddaa1 PM	174	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	175	},
964ddaa1 PM	176	{
	177	.hook = ipv4_confirm,
	178	.owner = THIS_MODULE,
57750a22	179	.pf = NFPROTO_IPV4,
6e23ae2a	180	.hooknum = NF_INET_POST_ROUTING,
964ddaa1 PM	181	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
	182	},
	183	{
	184	.hook = ipv4_confirm,
	185	.owner = THIS_MODULE,
57750a22	186	.pf = NFPROTO_IPV4,
6e23ae2a	187	.hooknum = NF_INET_LOCAL_IN,
964ddaa1 PM	188	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
964ddaa1 PM	189	},
9fb9cbb1 YK	190	};
9fb9cbb1 YK	191
a999e683 PM	192	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	193	static int log_invalid_proto_min = 0;
	194	static int log_invalid_proto_max = 255;
	195
	196	static ctl_table ip_ct_sysctl_table[] = {
	197	{
a999e683 PM	198	.procname = "ip_conntrack_max",
	199	.data = &nf_conntrack_max,
	200	.maxlen = sizeof(int),
	201	.mode = 0644,
6d9f239a	202	.proc_handler = proc_dointvec,
a999e683 PM	203	},
a999e683 PM	204	{
a999e683	205	.procname = "ip_conntrack_count",
49ac8713	206	.data = &init_net.ct.count,
a999e683 PM	207	.maxlen = sizeof(int),
a999e683 PM	208	.mode = 0444,
6d9f239a	209	.proc_handler = proc_dointvec,
a999e683 PM	210	},
a999e683 PM	211	{
a999e683	212	.procname = "ip_conntrack_buckets",
d696c7bd	213	.data = &init_net.ct.htable_size,
a999e683 PM	214	.maxlen = sizeof(unsigned int),
a999e683 PM	215	.mode = 0444,
6d9f239a	216	.proc_handler = proc_dointvec,
a999e683 PM	217	},
a999e683 PM	218	{
a999e683	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	},
a999e683 PM	225	{
a999e683	226	.procname = "ip_conntrack_log_invalid",
c2a2c7e0	227	.data = &init_net.ct.sysctl_log_invalid,
a999e683 PM	228	.maxlen = sizeof(unsigned int),
a999e683 PM	229	.mode = 0644,
6d9f239a	230	.proc_handler = proc_dointvec_minmax,
a999e683 PM	231	.extra1 = &log_invalid_proto_min,
	232	.extra2 = &log_invalid_proto_max,
	233	},
f8572d8f	234	{ }
a999e683 PM	235	};
	236	#endif /* CONFIG_SYSCTL && CONFIG_NF_CONNTRACK_PROC_COMPAT */
	237
9fb9cbb1 YK	238	/* Fast function for those who don't want to parse /proc (and I don't
	239	blame them). */
	240	/* Reversing the socket's dst/src point of view gives us the reply
	241	mapping. */
	242	static int
	243	getorigdst(struct sock sk, int optval, void __user user, int *len)
	244	{
32948588 JE	245	const struct inet_sock *inet = inet_sk(sk);
32948588 JE	246	const struct nf_conntrack_tuple_hash *h;
9fb9cbb1	247	struct nf_conntrack_tuple tuple;
e905a9ed	248
443a70d5	249	memset(&tuple, 0, sizeof(tuple));
c720c7e8 ED	250	tuple.src.u3.ip = inet->inet_rcv_saddr;
	251	tuple.src.u.tcp.port = inet->inet_sport;
	252	tuple.dst.u3.ip = inet->inet_daddr;
	253	tuple.dst.u.tcp.port = inet->inet_dport;
9fb9cbb1	254	tuple.src.l3num = PF_INET;
54981279	255	tuple.dst.protonum = sk->sk_protocol;
9fb9cbb1	256
54981279 RL	257	/* We only do TCP and SCTP at the moment: is there a better way? */
	258	if (sk->sk_protocol != IPPROTO_TCP && sk->sk_protocol != IPPROTO_SCTP) {
	259	pr_debug("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n");
9fb9cbb1 YK	260	return -ENOPROTOOPT;
	261	}
	262
	263	if ((unsigned int) *len < sizeof(struct sockaddr_in)) {
0d53778e PM	264	pr_debug("SO_ORIGINAL_DST: len %d not %Zu\n",
0d53778e PM	265	*len, sizeof(struct sockaddr_in));
9fb9cbb1 YK	266	return -EINVAL;
	267	}
	268
400dad39	269	h = nf_conntrack_find_get(sock_net(sk), &tuple);
9fb9cbb1 YK	270	if (h) {
	271	struct sockaddr_in sin;
	272	struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
	273
	274	sin.sin_family = AF_INET;
	275	sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	276	.tuple.dst.u.tcp.port;
	277	sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	278	.tuple.dst.u3.ip;
6c813c3f	279	memset(sin.sin_zero, 0, sizeof(sin.sin_zero));
9fb9cbb1	280
cffee385 HH	281	pr_debug("SO_ORIGINAL_DST: %pI4 %u\n",
cffee385 HH	282	&sin.sin_addr.s_addr, ntohs(sin.sin_port));
9fb9cbb1 YK	283	nf_ct_put(ct);
	284	if (copy_to_user(user, &sin, sizeof(sin)) != 0)
	285	return -EFAULT;
	286	else
	287	return 0;
	288	}
cffee385 HH	289	pr_debug("SO_ORIGINAL_DST: Can't find %pI4/%u-%pI4/%u.\n",
	290	&tuple.src.u3.ip, ntohs(tuple.src.u.tcp.port),
	291	&tuple.dst.u3.ip, ntohs(tuple.dst.u.tcp.port));
9fb9cbb1 YK	292	return -ENOENT;
	293	}
	294
e281db5c	295	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
c1d10adb PNA	296
	297	#include <linux/netfilter/nfnetlink.h>
	298	#include <linux/netfilter/nfnetlink_conntrack.h>
	299
fdf70832	300	static int ipv4_tuple_to_nlattr(struct sk_buff *skb,
c1d10adb PNA	301	const struct nf_conntrack_tuple *tuple)
c1d10adb PNA	302	{
77236b6e PM	303	NLA_PUT_BE32(skb, CTA_IP_V4_SRC, tuple->src.u3.ip);
77236b6e PM	304	NLA_PUT_BE32(skb, CTA_IP_V4_DST, tuple->dst.u3.ip);
c1d10adb PNA	305	return 0;
c1d10adb PNA	306
df6fb868	307	nla_put_failure:
c1d10adb PNA	308	return -1;
	309	}
	310
f73e924c PM	311	static const struct nla_policy ipv4_nla_policy[CTA_IP_MAX+1] = {
	312	[CTA_IP_V4_SRC] = { .type = NLA_U32 },
	313	[CTA_IP_V4_DST] = { .type = NLA_U32 },
c1d10adb PNA	314	};
c1d10adb PNA	315
fdf70832	316	static int ipv4_nlattr_to_tuple(struct nlattr *tb[],
c1d10adb PNA	317	struct nf_conntrack_tuple *t)
c1d10adb PNA	318	{
df6fb868	319	if (!tb[CTA_IP_V4_SRC] \|\| !tb[CTA_IP_V4_DST])
c1d10adb PNA	320	return -EINVAL;
c1d10adb PNA	321
77236b6e PM	322	t->src.u3.ip = nla_get_be32(tb[CTA_IP_V4_SRC]);
77236b6e PM	323	t->dst.u3.ip = nla_get_be32(tb[CTA_IP_V4_DST]);
c1d10adb PNA	324
	325	return 0;
	326	}
a400c30e HE	327
	328	static int ipv4_nlattr_tuple_size(void)
	329	{
	330	return nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1);
	331	}
c1d10adb PNA	332	#endif
c1d10adb PNA	333
9fb9cbb1 YK	334	static struct nf_sockopt_ops so_getorigdst = {
	335	.pf = PF_INET,
	336	.get_optmin = SO_ORIGINAL_DST,
	337	.get_optmax = SO_ORIGINAL_DST+1,
	338	.get = &getorigdst,
16fcec35	339	.owner = THIS_MODULE,
9fb9cbb1 YK	340	};
9fb9cbb1 YK	341
61075af5	342	struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
9fb9cbb1 YK	343	.l3proto = PF_INET,
	344	.name = "ipv4",
	345	.pkt_to_tuple = ipv4_pkt_to_tuple,
	346	.invert_tuple = ipv4_invert_tuple,
	347	.print_tuple = ipv4_print_tuple,
ffc30690	348	.get_l4proto = ipv4_get_l4proto,
e281db5c	349	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
fdf70832	350	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
a400c30e	351	.nlattr_tuple_size = ipv4_nlattr_tuple_size,
fdf70832	352	.nlattr_to_tuple = ipv4_nlattr_to_tuple,
f73e924c	353	.nla_policy = ipv4_nla_policy,
a999e683 PM	354	#endif
	355	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	356	.ctl_table_path = nf_net_ipv4_netfilter_sysctl_path,
	357	.ctl_table = ip_ct_sysctl_table,
c1d10adb	358	#endif
9fb9cbb1 YK	359	.me = THIS_MODULE,
	360	};
	361
fae718dd PM	362	module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint,
	363	&nf_conntrack_htable_size, 0600);
	364
32292a7f	365	MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET));
d2483dde	366	MODULE_ALIAS("ip_conntrack");
32292a7f PM	367	MODULE_LICENSE("GPL");
	368
	369	static int __init nf_conntrack_l3proto_ipv4_init(void)
9fb9cbb1 YK	370	{
	371	int ret = 0;
	372
32292a7f	373	need_conntrack();
73e4022f	374	nf_defrag_ipv4_enable();
9fb9cbb1 YK	375
	376	ret = nf_register_sockopt(&so_getorigdst);
	377	if (ret < 0) {
	378	printk(KERN_ERR "Unable to register netfilter socket option\n");
32292a7f	379	return ret;
9fb9cbb1 YK	380	}
9fb9cbb1 YK	381
605dcad6	382	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	383	if (ret < 0) {
	384	printk("nf_conntrack_ipv4: can't register tcp.\n");
	385	goto cleanup_sockopt;
	386	}
	387
605dcad6	388	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp4);
9fb9cbb1 YK	389	if (ret < 0) {
	390	printk("nf_conntrack_ipv4: can't register udp.\n");
	391	goto cleanup_tcp;
	392	}
	393
605dcad6	394	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmp);
9fb9cbb1 YK	395	if (ret < 0) {
	396	printk("nf_conntrack_ipv4: can't register icmp.\n");
	397	goto cleanup_udp;
	398	}
	399
	400	ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4);
	401	if (ret < 0) {
	402	printk("nf_conntrack_ipv4: can't register ipv4\n");
	403	goto cleanup_icmp;
	404	}
	405
964ddaa1 PM	406	ret = nf_register_hooks(ipv4_conntrack_ops,
964ddaa1 PM	407	ARRAY_SIZE(ipv4_conntrack_ops));
9fb9cbb1	408	if (ret < 0) {
964ddaa1	409	printk("nf_conntrack_ipv4: can't register hooks.\n");
9fb9cbb1 YK	410	goto cleanup_ipv4;
9fb9cbb1 YK	411	}
e4bd8bce PM	412	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	413	ret = nf_conntrack_ipv4_compat_init();
	414	if (ret < 0)
	415	goto cleanup_hooks;
	416	#endif
9fb9cbb1	417	return ret;
e4bd8bce PM	418	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
e4bd8bce PM	419	cleanup_hooks:
e905a9ed	420	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
e4bd8bce	421	#endif
9fb9cbb1 YK	422	cleanup_ipv4:
	423	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
	424	cleanup_icmp:
605dcad6	425	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
9fb9cbb1	426	cleanup_udp:
605dcad6	427	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
9fb9cbb1	428	cleanup_tcp:
605dcad6	429	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	430	cleanup_sockopt:
9fb9cbb1 YK	431	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	432	return ret;
	433	}
	434
65b4b4e8	435	static void __exit nf_conntrack_l3proto_ipv4_fini(void)
9fb9cbb1	436	{
32292a7f	437	synchronize_net();
e4bd8bce PM	438	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	439	nf_conntrack_ipv4_compat_fini();
	440	#endif
32292a7f PM	441	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
32292a7f PM	442	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
605dcad6 MJ	443	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
	444	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
	445	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
32292a7f	446	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	447	}
9fb9cbb1 YK	448
65b4b4e8 AM	449	module_init(nf_conntrack_l3proto_ipv4_init);
65b4b4e8 AM	450	module_exit(nf_conntrack_l3proto_ipv4_fini);
591e6206 PM	451
	452	void need_ipv4_conntrack(void)
	453	{
	454	return;
	455	}
	456	EXPORT_SYMBOL_GPL(need_ipv4_conntrack);