[linux-2.6-block.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))
			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)) {
		if (net_ratelimit())
			printk("ipt_hook: happy cracking.\n");
		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		= PF_INET,
		.hooknum	= NF_INET_PRE_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK,
	},
	{
		.hook		= ipv4_conntrack_local,
		.owner		= THIS_MODULE,
		.pf		= PF_INET,
		.hooknum	= NF_INET_LOCAL_OUT,
		.priority	= NF_IP_PRI_CONNTRACK,
	},
	{
		.hook		= ipv4_confirm,
		.owner		= THIS_MODULE,
		.pf		= PF_INET,
		.hooknum	= NF_INET_POST_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
	},
	{
		.hook		= ipv4_confirm,
		.owner		= THIS_MODULE,
		.pf		= PF_INET,
		.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 = IPPROTO_TCP;

	/* We only do TCP at the moment: is there a better way? */
	if (strcmp(sk->sk_prot->name, "TCP")) {
		pr_debug("SO_ORIGINAL_DST: Not a TCP 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;
}
#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_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);
	123	if (!seq_adjust \|\| !seq_adjust(skb, ct, ctinfo))
	124	return NF_DROP;
	125	}
	126	out:
	127	/* We've seen it coming out the other side: confirm it */
	128	return nf_conntrack_confirm(skb);
9fb9cbb1 YK	129	}
9fb9cbb1 YK	130
9fb9cbb1	131	static unsigned int ipv4_conntrack_in(unsigned int hooknum,
3db05fea	132	struct sk_buff *skb,
9fb9cbb1 YK	133	const struct net_device *in,
	134	const struct net_device *out,
	135	int (okfn)(struct sk_buff ))
	136	{
a702a65f	137	return nf_conntrack_in(dev_net(in), PF_INET, hooknum, skb);
9fb9cbb1 YK	138	}
	139
	140	static unsigned int ipv4_conntrack_local(unsigned int hooknum,
3db05fea	141	struct sk_buff *skb,
e905a9ed YH	142	const struct net_device *in,
	143	const struct net_device *out,
	144	int (okfn)(struct sk_buff ))
9fb9cbb1 YK	145	{
9fb9cbb1 YK	146	/* root is playing with raw sockets. */
3db05fea HX	147	if (skb->len < sizeof(struct iphdr) \|\|
3db05fea HX	148	ip_hdrlen(skb) < sizeof(struct iphdr)) {
9fb9cbb1 YK	149	if (net_ratelimit())
	150	printk("ipt_hook: happy cracking.\n");
	151	return NF_ACCEPT;
	152	}
a702a65f	153	return nf_conntrack_in(dev_net(out), PF_INET, hooknum, skb);
9fb9cbb1 YK	154	}
	155
	156	/* Connection tracking may drop packets, but never alters them, so
	157	make it the first hook. */
1999414a	158	static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
964ddaa1 PM	159	{
	160	.hook = ipv4_conntrack_in,
	161	.owner = THIS_MODULE,
	162	.pf = PF_INET,
6e23ae2a	163	.hooknum = NF_INET_PRE_ROUTING,
964ddaa1 PM	164	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	165	},
964ddaa1 PM	166	{
	167	.hook = ipv4_conntrack_local,
	168	.owner = THIS_MODULE,
	169	.pf = PF_INET,
6e23ae2a	170	.hooknum = NF_INET_LOCAL_OUT,
964ddaa1 PM	171	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	172	},
964ddaa1 PM	173	{
	174	.hook = ipv4_confirm,
	175	.owner = THIS_MODULE,
	176	.pf = PF_INET,
6e23ae2a	177	.hooknum = NF_INET_POST_ROUTING,
964ddaa1 PM	178	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
	179	},
	180	{
	181	.hook = ipv4_confirm,
	182	.owner = THIS_MODULE,
	183	.pf = PF_INET,
6e23ae2a	184	.hooknum = NF_INET_LOCAL_IN,
964ddaa1 PM	185	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
964ddaa1 PM	186	},
9fb9cbb1 YK	187	};
9fb9cbb1 YK	188
a999e683 PM	189	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	190	static int log_invalid_proto_min = 0;
	191	static int log_invalid_proto_max = 255;
	192
	193	static ctl_table ip_ct_sysctl_table[] = {
	194	{
	195	.ctl_name = NET_IPV4_NF_CONNTRACK_MAX,
	196	.procname = "ip_conntrack_max",
	197	.data = &nf_conntrack_max,
	198	.maxlen = sizeof(int),
	199	.mode = 0644,
6d9f239a	200	.proc_handler = proc_dointvec,
a999e683 PM	201	},
	202	{
	203	.ctl_name = NET_IPV4_NF_CONNTRACK_COUNT,
	204	.procname = "ip_conntrack_count",
49ac8713	205	.data = &init_net.ct.count,
a999e683 PM	206	.maxlen = sizeof(int),
a999e683 PM	207	.mode = 0444,
6d9f239a	208	.proc_handler = proc_dointvec,
a999e683 PM	209	},
	210	{
	211	.ctl_name = NET_IPV4_NF_CONNTRACK_BUCKETS,
	212	.procname = "ip_conntrack_buckets",
	213	.data = &nf_conntrack_htable_size,
	214	.maxlen = sizeof(unsigned int),
	215	.mode = 0444,
6d9f239a	216	.proc_handler = proc_dointvec,
a999e683 PM	217	},
	218	{
	219	.ctl_name = NET_IPV4_NF_CONNTRACK_CHECKSUM,
	220	.procname = "ip_conntrack_checksum",
c04d0552	221	.data = &init_net.ct.sysctl_checksum,
a999e683 PM	222	.maxlen = sizeof(int),
a999e683 PM	223	.mode = 0644,
6d9f239a	224	.proc_handler = proc_dointvec,
a999e683 PM	225	},
	226	{
	227	.ctl_name = NET_IPV4_NF_CONNTRACK_LOG_INVALID,
	228	.procname = "ip_conntrack_log_invalid",
c2a2c7e0	229	.data = &init_net.ct.sysctl_log_invalid,
a999e683 PM	230	.maxlen = sizeof(unsigned int),
a999e683 PM	231	.mode = 0644,
6d9f239a AD	232	.proc_handler = proc_dointvec_minmax,
6d9f239a AD	233	.strategy = sysctl_intvec,
a999e683 PM	234	.extra1 = &log_invalid_proto_min,
	235	.extra2 = &log_invalid_proto_max,
	236	},
	237	{
	238	.ctl_name = 0
	239	}
	240	};
	241	#endif /* CONFIG_SYSCTL && CONFIG_NF_CONNTRACK_PROC_COMPAT */
	242
9fb9cbb1 YK	243	/* Fast function for those who don't want to parse /proc (and I don't
	244	blame them). */
	245	/* Reversing the socket's dst/src point of view gives us the reply
	246	mapping. */
	247	static int
	248	getorigdst(struct sock sk, int optval, void __user user, int *len)
	249	{
32948588 JE	250	const struct inet_sock *inet = inet_sk(sk);
32948588 JE	251	const struct nf_conntrack_tuple_hash *h;
9fb9cbb1	252	struct nf_conntrack_tuple tuple;
e905a9ed	253
443a70d5	254	memset(&tuple, 0, sizeof(tuple));
9fb9cbb1 YK	255	tuple.src.u3.ip = inet->rcv_saddr;
	256	tuple.src.u.tcp.port = inet->sport;
	257	tuple.dst.u3.ip = inet->daddr;
	258	tuple.dst.u.tcp.port = inet->dport;
	259	tuple.src.l3num = PF_INET;
	260	tuple.dst.protonum = IPPROTO_TCP;
	261
	262	/* We only do TCP at the moment: is there a better way? */
	263	if (strcmp(sk->sk_prot->name, "TCP")) {
0d53778e	264	pr_debug("SO_ORIGINAL_DST: Not a TCP socket\n");
9fb9cbb1 YK	265	return -ENOPROTOOPT;
	266	}
	267
	268	if ((unsigned int) *len < sizeof(struct sockaddr_in)) {
0d53778e PM	269	pr_debug("SO_ORIGINAL_DST: len %d not %Zu\n",
0d53778e PM	270	*len, sizeof(struct sockaddr_in));
9fb9cbb1 YK	271	return -EINVAL;
	272	}
	273
400dad39	274	h = nf_conntrack_find_get(sock_net(sk), &tuple);
9fb9cbb1 YK	275	if (h) {
	276	struct sockaddr_in sin;
	277	struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
	278
	279	sin.sin_family = AF_INET;
	280	sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	281	.tuple.dst.u.tcp.port;
	282	sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	283	.tuple.dst.u3.ip;
6c813c3f	284	memset(sin.sin_zero, 0, sizeof(sin.sin_zero));
9fb9cbb1	285
cffee385 HH	286	pr_debug("SO_ORIGINAL_DST: %pI4 %u\n",
cffee385 HH	287	&sin.sin_addr.s_addr, ntohs(sin.sin_port));
9fb9cbb1 YK	288	nf_ct_put(ct);
	289	if (copy_to_user(user, &sin, sizeof(sin)) != 0)
	290	return -EFAULT;
	291	else
	292	return 0;
	293	}
cffee385 HH	294	pr_debug("SO_ORIGINAL_DST: Can't find %pI4/%u-%pI4/%u.\n",
	295	&tuple.src.u3.ip, ntohs(tuple.src.u.tcp.port),
	296	&tuple.dst.u3.ip, ntohs(tuple.dst.u.tcp.port));
9fb9cbb1 YK	297	return -ENOENT;
	298	}
	299
e281db5c	300	#if defined(CONFIG_NF_CT_NETLINK) \|\| defined(CONFIG_NF_CT_NETLINK_MODULE)
c1d10adb PNA	301
	302	#include <linux/netfilter/nfnetlink.h>
	303	#include <linux/netfilter/nfnetlink_conntrack.h>
	304
fdf70832	305	static int ipv4_tuple_to_nlattr(struct sk_buff *skb,
c1d10adb PNA	306	const struct nf_conntrack_tuple *tuple)
c1d10adb PNA	307	{
77236b6e PM	308	NLA_PUT_BE32(skb, CTA_IP_V4_SRC, tuple->src.u3.ip);
77236b6e PM	309	NLA_PUT_BE32(skb, CTA_IP_V4_DST, tuple->dst.u3.ip);
c1d10adb PNA	310	return 0;
c1d10adb PNA	311
df6fb868	312	nla_put_failure:
c1d10adb PNA	313	return -1;
	314	}
	315
f73e924c PM	316	static const struct nla_policy ipv4_nla_policy[CTA_IP_MAX+1] = {
	317	[CTA_IP_V4_SRC] = { .type = NLA_U32 },
	318	[CTA_IP_V4_DST] = { .type = NLA_U32 },
c1d10adb PNA	319	};
c1d10adb PNA	320
fdf70832	321	static int ipv4_nlattr_to_tuple(struct nlattr *tb[],
c1d10adb PNA	322	struct nf_conntrack_tuple *t)
c1d10adb PNA	323	{
df6fb868	324	if (!tb[CTA_IP_V4_SRC] \|\| !tb[CTA_IP_V4_DST])
c1d10adb PNA	325	return -EINVAL;
c1d10adb PNA	326
77236b6e PM	327	t->src.u3.ip = nla_get_be32(tb[CTA_IP_V4_SRC]);
77236b6e PM	328	t->dst.u3.ip = nla_get_be32(tb[CTA_IP_V4_DST]);
c1d10adb PNA	329
	330	return 0;
	331	}
	332	#endif
	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 PM	350	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
fdf70832 PM	351	.nlattr_to_tuple = ipv4_nlattr_to_tuple,
f73e924c	352	.nla_policy = ipv4_nla_policy,
a999e683 PM	353	#endif
	354	#if defined(CONFIG_SYSCTL) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	355	.ctl_table_path = nf_net_ipv4_netfilter_sysctl_path,
	356	.ctl_table = ip_ct_sysctl_table,
c1d10adb	357	#endif
9fb9cbb1 YK	358	.me = THIS_MODULE,
	359	};
	360
fae718dd PM	361	module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint,
	362	&nf_conntrack_htable_size, 0600);
	363
32292a7f	364	MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET));
d2483dde	365	MODULE_ALIAS("ip_conntrack");
32292a7f PM	366	MODULE_LICENSE("GPL");
	367
	368	static int __init nf_conntrack_l3proto_ipv4_init(void)
9fb9cbb1 YK	369	{
	370	int ret = 0;
	371
32292a7f	372	need_conntrack();
73e4022f	373	nf_defrag_ipv4_enable();
9fb9cbb1 YK	374
	375	ret = nf_register_sockopt(&so_getorigdst);
	376	if (ret < 0) {
	377	printk(KERN_ERR "Unable to register netfilter socket option\n");
32292a7f	378	return ret;
9fb9cbb1 YK	379	}
9fb9cbb1 YK	380
605dcad6	381	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	382	if (ret < 0) {
	383	printk("nf_conntrack_ipv4: can't register tcp.\n");
	384	goto cleanup_sockopt;
	385	}
	386
605dcad6	387	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_udp4);
9fb9cbb1 YK	388	if (ret < 0) {
	389	printk("nf_conntrack_ipv4: can't register udp.\n");
	390	goto cleanup_tcp;
	391	}
	392
605dcad6	393	ret = nf_conntrack_l4proto_register(&nf_conntrack_l4proto_icmp);
9fb9cbb1 YK	394	if (ret < 0) {
	395	printk("nf_conntrack_ipv4: can't register icmp.\n");
	396	goto cleanup_udp;
	397	}
	398
	399	ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv4);
	400	if (ret < 0) {
	401	printk("nf_conntrack_ipv4: can't register ipv4\n");
	402	goto cleanup_icmp;
	403	}
	404
964ddaa1 PM	405	ret = nf_register_hooks(ipv4_conntrack_ops,
964ddaa1 PM	406	ARRAY_SIZE(ipv4_conntrack_ops));
9fb9cbb1	407	if (ret < 0) {
964ddaa1	408	printk("nf_conntrack_ipv4: can't register hooks.\n");
9fb9cbb1 YK	409	goto cleanup_ipv4;
9fb9cbb1 YK	410	}
e4bd8bce PM	411	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	412	ret = nf_conntrack_ipv4_compat_init();
	413	if (ret < 0)
	414	goto cleanup_hooks;
	415	#endif
9fb9cbb1	416	return ret;
e4bd8bce PM	417	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
e4bd8bce PM	418	cleanup_hooks:
e905a9ed	419	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
e4bd8bce	420	#endif
9fb9cbb1 YK	421	cleanup_ipv4:
	422	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
	423	cleanup_icmp:
605dcad6	424	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
9fb9cbb1	425	cleanup_udp:
605dcad6	426	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
9fb9cbb1	427	cleanup_tcp:
605dcad6	428	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
9fb9cbb1 YK	429	cleanup_sockopt:
9fb9cbb1 YK	430	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	431	return ret;
	432	}
	433
65b4b4e8	434	static void __exit nf_conntrack_l3proto_ipv4_fini(void)
9fb9cbb1	435	{
32292a7f	436	synchronize_net();
e4bd8bce PM	437	#if defined(CONFIG_PROC_FS) && defined(CONFIG_NF_CONNTRACK_PROC_COMPAT)
	438	nf_conntrack_ipv4_compat_fini();
	439	#endif
32292a7f PM	440	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
32292a7f PM	441	nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
605dcad6 MJ	442	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_icmp);
	443	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_udp4);
	444	nf_conntrack_l4proto_unregister(&nf_conntrack_l4proto_tcp4);
32292a7f	445	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	446	}
9fb9cbb1 YK	447
65b4b4e8 AM	448	module_init(nf_conntrack_l3proto_ipv4_init);
65b4b4e8 AM	449	module_exit(nf_conntrack_l3proto_ipv4_fini);
591e6206 PM	450
	451	void need_ipv4_conntrack(void)
	452	{
	453	return;
	454	}
	455	EXPORT_SYMBOL_GPL(need_ipv4_conntrack);