[linux-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>
 * (C) 2006-2012 Patrick McHardy <kaber@trash.net>
 *
 * 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/nf_conntrack_seqadj.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>

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 void ipv4_print_tuple(struct seq_file *s,
			    const struct nf_conntrack_tuple *tuple)
{
	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;

	/* Check bogus IP headers */
	if (*dataoff > skb->len) {
		pr_debug("nf_conntrack_ipv4: bogus IPv4 packet: "
			 "nhoff %u, ihl %u, skblen %u\n",
			 nhoff, iph->ihl << 2, skb->len);
		return -NF_ACCEPT;
	}

	return NF_ACCEPT;
}

static unsigned int ipv4_helper(void *priv,
				struct sk_buff *skb,
				const struct nf_hook_state *state)
{
	struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;
	const struct nf_conn_help *help;
	const struct nf_conntrack_helper *helper;

	/* 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)
		return NF_ACCEPT;

	help = nfct_help(ct);
	if (!help)
		return NF_ACCEPT;

	/* rcu_read_lock()ed by nf_hook_thresh */
	helper = rcu_dereference(help->helper);
	if (!helper)
		return NF_ACCEPT;

	return helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
			    ct, ctinfo);
}

static unsigned int ipv4_confirm(void *priv,
				 struct sk_buff *skb,
				 const struct nf_hook_state *state)
{
	struct nf_conn *ct;
	enum ip_conntrack_info ctinfo;

	ct = nf_ct_get(skb, &ctinfo);
	if (!ct || ctinfo == IP_CT_RELATED_REPLY)
		goto out;

	/* adjust seqs for loopback traffic only in outgoing direction */
	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
	    !nf_is_loopback_packet(skb)) {
		if (!nf_ct_seq_adjust(skb, ct, ctinfo, ip_hdrlen(skb))) {
			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(void *priv,
				      struct sk_buff *skb,
				      const struct nf_hook_state *state)
{
	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
}

static unsigned int ipv4_conntrack_local(void *priv,
					 struct sk_buff *skb,
					 const struct nf_hook_state *state)
{
	/* 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(state->net, PF_INET, state->hook, 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,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_PRE_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK,
	},
	{
		.hook		= ipv4_conntrack_local,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_LOCAL_OUT,
		.priority	= NF_IP_PRI_CONNTRACK,
	},
	{
		.hook		= ipv4_helper,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_POST_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK_HELPER,
	},
	{
		.hook		= ipv4_confirm,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_POST_ROUTING,
		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
	},
	{
		.hook		= ipv4_helper,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_LOCAL_IN,
		.priority	= NF_IP_PRI_CONNTRACK_HELPER,
	},
	{
		.hook		= ipv4_confirm,
		.pf		= NFPROTO_IPV4,
		.hooknum	= NF_INET_LOCAL_IN,
		.priority	= NF_IP_PRI_CONNTRACK_CONFIRM,
	},
};

/* 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_zone_dflt, &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 IS_ENABLED(CONFIG_NF_CT_NETLINK)

#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)
{
	if (nla_put_in_addr(skb, CTA_IP_V4_SRC, tuple->src.u3.ip) ||
	    nla_put_in_addr(skb, CTA_IP_V4_DST, tuple->dst.u3.ip))
		goto nla_put_failure;
	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_in_addr(tb[CTA_IP_V4_SRC]);
	t->dst.u3.ip = nla_get_in_addr(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,
};

static int ipv4_init_net(struct net *net)
{
	return 0;
}

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 IS_ENABLED(CONFIG_NF_CT_NETLINK)
	.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
	.init_net	 = ipv4_init_net,
	.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 struct nf_conntrack_l4proto *builtin_l4proto4[] = {
	&nf_conntrack_l4proto_tcp4,
	&nf_conntrack_l4proto_udp4,
	&nf_conntrack_l4proto_icmp,
#ifdef CONFIG_NF_CT_PROTO_DCCP
	&nf_conntrack_l4proto_dccp4,
#endif
#ifdef CONFIG_NF_CT_PROTO_SCTP
	&nf_conntrack_l4proto_sctp4,
#endif
};

static int ipv4_net_init(struct net *net)
{
	int ret = 0;

	ret = nf_ct_l4proto_pernet_register(net, builtin_l4proto4,
					    ARRAY_SIZE(builtin_l4proto4));
	if (ret < 0)
		return ret;
	ret = nf_ct_l3proto_pernet_register(net, &nf_conntrack_l3proto_ipv4);
	if (ret < 0) {
		pr_err("nf_conntrack_ipv4: pernet registration failed\n");
		nf_ct_l4proto_pernet_unregister(net, builtin_l4proto4,
						ARRAY_SIZE(builtin_l4proto4));
	}
	return ret;
}

static void ipv4_net_exit(struct net *net)
{
	nf_ct_l3proto_pernet_unregister(net, &nf_conntrack_l3proto_ipv4);
	nf_ct_l4proto_pernet_unregister(net, builtin_l4proto4,
					ARRAY_SIZE(builtin_l4proto4));
}

static struct pernet_operations ipv4_net_ops = {
	.init = ipv4_net_init,
	.exit = ipv4_net_exit,
};

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) {
		pr_err("Unable to register netfilter socket option\n");
		return ret;
	}

	ret = register_pernet_subsys(&ipv4_net_ops);
	if (ret < 0) {
		pr_err("nf_conntrack_ipv4: can't register pernet ops\n");
		goto cleanup_sockopt;
	}

	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_pernet;
	}

	ret = nf_ct_l4proto_register(builtin_l4proto4,
				     ARRAY_SIZE(builtin_l4proto4));
	if (ret < 0)
		goto cleanup_hooks;

	ret = nf_ct_l3proto_register(&nf_conntrack_l3proto_ipv4);
	if (ret < 0) {
		pr_err("nf_conntrack_ipv4: can't register ipv4 proto.\n");
		goto cleanup_l4proto;
	}

	return ret;
cleanup_l4proto:
	nf_ct_l4proto_unregister(builtin_l4proto4,
				 ARRAY_SIZE(builtin_l4proto4));
 cleanup_hooks:
	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
 cleanup_pernet:
	unregister_pernet_subsys(&ipv4_net_ops);
 cleanup_sockopt:
	nf_unregister_sockopt(&so_getorigdst);
	return ret;
}

static void __exit nf_conntrack_l3proto_ipv4_fini(void)
{
	synchronize_net();
	nf_ct_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
	nf_ct_l4proto_unregister(builtin_l4proto4,
				 ARRAY_SIZE(builtin_l4proto4));
	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
	unregister_pernet_subsys(&ipv4_net_ops);
	nf_unregister_sockopt(&so_getorigdst);
}

module_init(nf_conntrack_l3proto_ipv4_init);
module_exit(nf_conntrack_l3proto_ipv4_fini);
Commit	Line	Data
73e4022f	1
9fb9cbb1 YK	2	/* (C) 1999-2001 Paul `Rusty' Russell
9fb9cbb1 YK	3	* (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org>
f229f6ce	4	* (C) 2006-2012 Patrick McHardy <kaber@trash.net>
9fb9cbb1 YK	5	*
	6	* This program is free software; you can redistribute it and/or modify
	7	* it under the terms of the GNU General Public License version 2 as
	8	* published by the Free Software Foundation.
9fb9cbb1 YK	9	*/
9fb9cbb1 YK	10
9fb9cbb1 YK	11	#include <linux/types.h>
	12	#include <linux/ip.h>
	13	#include <linux/netfilter.h>
	14	#include <linux/module.h>
	15	#include <linux/skbuff.h>
	16	#include <linux/icmp.h>
	17	#include <linux/sysctl.h>
0ae2cfe7	18	#include <net/route.h>
9fb9cbb1 YK	19	#include <net/ip.h>
	20
	21	#include <linux/netfilter_ipv4.h>
	22	#include <net/netfilter/nf_conntrack.h>
	23	#include <net/netfilter/nf_conntrack_helper.h>
605dcad6	24	#include <net/netfilter/nf_conntrack_l4proto.h>
9fb9cbb1	25	#include <net/netfilter/nf_conntrack_l3proto.h>
5d0aa2cc	26	#include <net/netfilter/nf_conntrack_zones.h>
9fb9cbb1	27	#include <net/netfilter/nf_conntrack_core.h>
41d73ec0	28	#include <net/netfilter/nf_conntrack_seqadj.h>
9fb9cbb1	29	#include <net/netfilter/ipv4/nf_conntrack_ipv4.h>
dd13b010	30	#include <net/netfilter/nf_nat_helper.h>
73e4022f	31	#include <net/netfilter/ipv4/nf_defrag_ipv4.h>
74f7a655	32	#include <net/netfilter/nf_log.h>
dd13b010	33
8ce8439a JE	34	static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff,
8ce8439a JE	35	struct nf_conntrack_tuple *tuple)
9fb9cbb1	36	{
32948588 JE	37	const __be32 *ap;
32948588 JE	38	__be32 _addrs[2];
9fb9cbb1 YK	39	ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr),
	40	sizeof(u_int32_t) * 2, _addrs);
	41	if (ap == NULL)
8ce8439a	42	return false;
9fb9cbb1 YK	43
	44	tuple->src.u3.ip = ap[0];
	45	tuple->dst.u3.ip = ap[1];
	46
8ce8439a	47	return true;
9fb9cbb1 YK	48	}
9fb9cbb1 YK	49
8ce8439a JE	50	static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple,
8ce8439a JE	51	const struct nf_conntrack_tuple *orig)
9fb9cbb1 YK	52	{
	53	tuple->src.u3.ip = orig->dst.u3.ip;
	54	tuple->dst.u3.ip = orig->src.u3.ip;
	55
8ce8439a	56	return true;
9fb9cbb1 YK	57	}
9fb9cbb1 YK	58
824f1fbe	59	static void ipv4_print_tuple(struct seq_file *s,
9fb9cbb1 YK	60	const struct nf_conntrack_tuple *tuple)
9fb9cbb1 YK	61	{
824f1fbe JP	62	seq_printf(s, "src=%pI4 dst=%pI4 ",
824f1fbe JP	63	&tuple->src.u3.ip, &tuple->dst.u3.ip);
9fb9cbb1 YK	64	}
9fb9cbb1 YK	65
ffc30690 YK	66	static int ipv4_get_l4proto(const struct sk_buff *skb, unsigned int nhoff,
ffc30690 YK	67	unsigned int dataoff, u_int8_t protonum)
9fb9cbb1	68	{
32948588 JE	69	const struct iphdr *iph;
32948588 JE	70	struct iphdr _iph;
ffc30690 YK	71
	72	iph = skb_header_pointer(skb, nhoff, sizeof(_iph), &_iph);
	73	if (iph == NULL)
8430eac2	74	return -NF_ACCEPT;
ffc30690	75
0fb96701 PM	76	/* Conntrack defragments packets, we might still see fragments
	77	* inside ICMP packets though. */
	78	if (iph->frag_off & htons(IP_OFFSET))
8430eac2	79	return -NF_ACCEPT;
9fb9cbb1	80
ffc30690 YK	81	*dataoff = nhoff + (iph->ihl << 2);
ffc30690 YK	82	*protonum = iph->protocol;
9fb9cbb1	83
07153c6e JK	84	/* Check bogus IP headers */
	85	if (*dataoff > skb->len) {
	86	pr_debug("nf_conntrack_ipv4: bogus IPv4 packet: "
	87	"nhoff %u, ihl %u, skblen %u\n",
	88	nhoff, iph->ihl << 2, skb->len);
	89	return -NF_ACCEPT;
	90	}
	91
9fb9cbb1 YK	92	return NF_ACCEPT;
	93	}
	94
06198b34	95	static unsigned int ipv4_helper(void *priv,
12f7a505	96	struct sk_buff *skb,
238e54c9	97	const struct nf_hook_state *state)
9fb9cbb1 YK	98	{
	99	struct nf_conn *ct;
	100	enum ip_conntrack_info ctinfo;
32948588 JE	101	const struct nf_conn_help *help;
32948588 JE	102	const struct nf_conntrack_helper *helper;
9fb9cbb1 YK	103
9fb9cbb1 YK	104	/* This is where we call the helper: as the packet goes out. */
3db05fea	105	ct = nf_ct_get(skb, &ctinfo);
fb048833	106	if (!ct \|\| ctinfo == IP_CT_RELATED_REPLY)
12f7a505	107	return NF_ACCEPT;
dc808fe2 HW	108
dc808fe2 HW	109	help = nfct_help(ct);
3c158f7f	110	if (!help)
12f7a505	111	return NF_ACCEPT;
dd13b010	112
e2361cb9	113	/* rcu_read_lock()ed by nf_hook_thresh */
3c158f7f PM	114	helper = rcu_dereference(help->helper);
3c158f7f PM	115	if (!helper)
12f7a505	116	return NF_ACCEPT;
dd13b010	117
b20ab9cc PNA	118	return helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb),
b20ab9cc PNA	119	ct, ctinfo);
12f7a505 PNA	120	}
12f7a505 PNA	121
06198b34	122	static unsigned int ipv4_confirm(void *priv,
12f7a505	123	struct sk_buff *skb,
238e54c9	124	const struct nf_hook_state *state)
12f7a505 PNA	125	{
	126	struct nf_conn *ct;
	127	enum ip_conntrack_info ctinfo;
	128
	129	ct = nf_ct_get(skb, &ctinfo);
	130	if (!ct \|\| ctinfo == IP_CT_RELATED_REPLY)
	131	goto out;
dd13b010	132
42c1edd3 JA	133	/* adjust seqs for loopback traffic only in outgoing direction */
	134	if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status) &&
	135	!nf_is_loopback_packet(skb)) {
41d73ec0	136	if (!nf_ct_seq_adjust(skb, ct, ctinfo, ip_hdrlen(skb))) {
1db7a748	137	NF_CT_STAT_INC_ATOMIC(nf_ct_net(ct), drop);
dd13b010	138	return NF_DROP;
1db7a748	139	}
dd13b010 PM	140	}
	141	out:
	142	/* We've seen it coming out the other side: confirm it */
	143	return nf_conntrack_confirm(skb);
9fb9cbb1 YK	144	}
9fb9cbb1 YK	145
06198b34	146	static unsigned int ipv4_conntrack_in(void *priv,
3db05fea	147	struct sk_buff *skb,
238e54c9	148	const struct nf_hook_state *state)
9fb9cbb1	149	{
082a758f	150	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
9fb9cbb1 YK	151	}
9fb9cbb1 YK	152
06198b34	153	static unsigned int ipv4_conntrack_local(void *priv,
3db05fea	154	struct sk_buff *skb,
238e54c9	155	const struct nf_hook_state *state)
9fb9cbb1 YK	156	{
9fb9cbb1 YK	157	/* root is playing with raw sockets. */
3db05fea	158	if (skb->len < sizeof(struct iphdr) \|\|
88843104	159	ip_hdrlen(skb) < sizeof(struct iphdr))
9fb9cbb1	160	return NF_ACCEPT;
082a758f	161	return nf_conntrack_in(state->net, PF_INET, state->hook, skb);
9fb9cbb1 YK	162	}
	163
	164	/* Connection tracking may drop packets, but never alters them, so
	165	make it the first hook. */
1999414a	166	static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = {
964ddaa1 PM	167	{
964ddaa1 PM	168	.hook = ipv4_conntrack_in,
57750a22	169	.pf = NFPROTO_IPV4,
6e23ae2a	170	.hooknum = NF_INET_PRE_ROUTING,
964ddaa1 PM	171	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	172	},
964ddaa1 PM	173	{
964ddaa1 PM	174	.hook = ipv4_conntrack_local,
57750a22	175	.pf = NFPROTO_IPV4,
6e23ae2a	176	.hooknum = NF_INET_LOCAL_OUT,
964ddaa1 PM	177	.priority = NF_IP_PRI_CONNTRACK,
964ddaa1 PM	178	},
12f7a505 PNA	179	{
12f7a505 PNA	180	.hook = ipv4_helper,
12f7a505 PNA	181	.pf = NFPROTO_IPV4,
	182	.hooknum = NF_INET_POST_ROUTING,
	183	.priority = NF_IP_PRI_CONNTRACK_HELPER,
	184	},
964ddaa1 PM	185	{
964ddaa1 PM	186	.hook = ipv4_confirm,
57750a22	187	.pf = NFPROTO_IPV4,
6e23ae2a	188	.hooknum = NF_INET_POST_ROUTING,
964ddaa1 PM	189	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
964ddaa1 PM	190	},
12f7a505 PNA	191	{
12f7a505 PNA	192	.hook = ipv4_helper,
12f7a505 PNA	193	.pf = NFPROTO_IPV4,
	194	.hooknum = NF_INET_LOCAL_IN,
	195	.priority = NF_IP_PRI_CONNTRACK_HELPER,
	196	},
964ddaa1 PM	197	{
964ddaa1 PM	198	.hook = ipv4_confirm,
57750a22	199	.pf = NFPROTO_IPV4,
6e23ae2a	200	.hooknum = NF_INET_LOCAL_IN,
964ddaa1 PM	201	.priority = NF_IP_PRI_CONNTRACK_CONFIRM,
964ddaa1 PM	202	},
9fb9cbb1 YK	203	};
9fb9cbb1 YK	204
9fb9cbb1 YK	205	/* Fast function for those who don't want to parse /proc (and I don't
	206	blame them). */
	207	/* Reversing the socket's dst/src point of view gives us the reply
	208	mapping. */
	209	static int
	210	getorigdst(struct sock sk, int optval, void __user user, int *len)
	211	{
32948588 JE	212	const struct inet_sock *inet = inet_sk(sk);
32948588 JE	213	const struct nf_conntrack_tuple_hash *h;
9fb9cbb1	214	struct nf_conntrack_tuple tuple;
e905a9ed	215
443a70d5	216	memset(&tuple, 0, sizeof(tuple));
c720c7e8 ED	217	tuple.src.u3.ip = inet->inet_rcv_saddr;
	218	tuple.src.u.tcp.port = inet->inet_sport;
	219	tuple.dst.u3.ip = inet->inet_daddr;
	220	tuple.dst.u.tcp.port = inet->inet_dport;
9fb9cbb1	221	tuple.src.l3num = PF_INET;
54981279	222	tuple.dst.protonum = sk->sk_protocol;
9fb9cbb1	223
54981279 RL	224	/* We only do TCP and SCTP at the moment: is there a better way? */
	225	if (sk->sk_protocol != IPPROTO_TCP && sk->sk_protocol != IPPROTO_SCTP) {
	226	pr_debug("SO_ORIGINAL_DST: Not a TCP/SCTP socket\n");
9fb9cbb1 YK	227	return -ENOPROTOOPT;
	228	}
	229
	230	if ((unsigned int) *len < sizeof(struct sockaddr_in)) {
0d53778e PM	231	pr_debug("SO_ORIGINAL_DST: len %d not %Zu\n",
0d53778e PM	232	*len, sizeof(struct sockaddr_in));
9fb9cbb1 YK	233	return -EINVAL;
	234	}
	235
308ac914	236	h = nf_conntrack_find_get(sock_net(sk), &nf_ct_zone_dflt, &tuple);
9fb9cbb1 YK	237	if (h) {
	238	struct sockaddr_in sin;
	239	struct nf_conn *ct = nf_ct_tuplehash_to_ctrack(h);
	240
	241	sin.sin_family = AF_INET;
	242	sin.sin_port = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	243	.tuple.dst.u.tcp.port;
	244	sin.sin_addr.s_addr = ct->tuplehash[IP_CT_DIR_ORIGINAL]
	245	.tuple.dst.u3.ip;
6c813c3f	246	memset(sin.sin_zero, 0, sizeof(sin.sin_zero));
9fb9cbb1	247
cffee385 HH	248	pr_debug("SO_ORIGINAL_DST: %pI4 %u\n",
cffee385 HH	249	&sin.sin_addr.s_addr, ntohs(sin.sin_port));
9fb9cbb1 YK	250	nf_ct_put(ct);
	251	if (copy_to_user(user, &sin, sizeof(sin)) != 0)
	252	return -EFAULT;
	253	else
	254	return 0;
	255	}
cffee385 HH	256	pr_debug("SO_ORIGINAL_DST: Can't find %pI4/%u-%pI4/%u.\n",
	257	&tuple.src.u3.ip, ntohs(tuple.src.u.tcp.port),
	258	&tuple.dst.u3.ip, ntohs(tuple.dst.u.tcp.port));
9fb9cbb1 YK	259	return -ENOENT;
	260	}
	261
24de3d37	262	#if IS_ENABLED(CONFIG_NF_CT_NETLINK)
c1d10adb PNA	263
	264	#include <linux/netfilter/nfnetlink.h>
	265	#include <linux/netfilter/nfnetlink_conntrack.h>
	266
fdf70832	267	static int ipv4_tuple_to_nlattr(struct sk_buff *skb,
c1d10adb PNA	268	const struct nf_conntrack_tuple *tuple)
c1d10adb PNA	269	{
930345ea JB	270	if (nla_put_in_addr(skb, CTA_IP_V4_SRC, tuple->src.u3.ip) \|\|
930345ea JB	271	nla_put_in_addr(skb, CTA_IP_V4_DST, tuple->dst.u3.ip))
d317e4f6	272	goto nla_put_failure;
c1d10adb PNA	273	return 0;
c1d10adb PNA	274
df6fb868	275	nla_put_failure:
c1d10adb PNA	276	return -1;
	277	}
	278
f73e924c PM	279	static const struct nla_policy ipv4_nla_policy[CTA_IP_MAX+1] = {
	280	[CTA_IP_V4_SRC] = { .type = NLA_U32 },
	281	[CTA_IP_V4_DST] = { .type = NLA_U32 },
c1d10adb PNA	282	};
c1d10adb PNA	283
fdf70832	284	static int ipv4_nlattr_to_tuple(struct nlattr *tb[],
c1d10adb PNA	285	struct nf_conntrack_tuple *t)
c1d10adb PNA	286	{
df6fb868	287	if (!tb[CTA_IP_V4_SRC] \|\| !tb[CTA_IP_V4_DST])
c1d10adb PNA	288	return -EINVAL;
c1d10adb PNA	289
67b61f6c JB	290	t->src.u3.ip = nla_get_in_addr(tb[CTA_IP_V4_SRC]);
67b61f6c JB	291	t->dst.u3.ip = nla_get_in_addr(tb[CTA_IP_V4_DST]);
c1d10adb PNA	292
	293	return 0;
	294	}
a400c30e HE	295
	296	static int ipv4_nlattr_tuple_size(void)
	297	{
	298	return nla_policy_len(ipv4_nla_policy, CTA_IP_MAX + 1);
	299	}
c1d10adb PNA	300	#endif
c1d10adb PNA	301
9fb9cbb1 YK	302	static struct nf_sockopt_ops so_getorigdst = {
	303	.pf = PF_INET,
	304	.get_optmin = SO_ORIGINAL_DST,
	305	.get_optmax = SO_ORIGINAL_DST+1,
5bd3a76f	306	.get = getorigdst,
16fcec35	307	.owner = THIS_MODULE,
9fb9cbb1 YK	308	};
9fb9cbb1 YK	309
3ea04dd3 G	310	static int ipv4_init_net(struct net *net)
3ea04dd3 G	311	{
3ea04dd3 G	312	return 0;
	313	}
	314
61075af5	315	struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv4 __read_mostly = {
9fb9cbb1 YK	316	.l3proto = PF_INET,
	317	.name = "ipv4",
	318	.pkt_to_tuple = ipv4_pkt_to_tuple,
	319	.invert_tuple = ipv4_invert_tuple,
	320	.print_tuple = ipv4_print_tuple,
ffc30690	321	.get_l4proto = ipv4_get_l4proto,
24de3d37	322	#if IS_ENABLED(CONFIG_NF_CT_NETLINK)
fdf70832	323	.tuple_to_nlattr = ipv4_tuple_to_nlattr,
a400c30e	324	.nlattr_tuple_size = ipv4_nlattr_tuple_size,
fdf70832	325	.nlattr_to_tuple = ipv4_nlattr_to_tuple,
f73e924c	326	.nla_policy = ipv4_nla_policy,
c1d10adb	327	#endif
3ea04dd3	328	.init_net = ipv4_init_net,
9fb9cbb1 YK	329	.me = THIS_MODULE,
	330	};
	331
fae718dd PM	332	module_param_call(hashsize, nf_conntrack_set_hashsize, param_get_uint,
	333	&nf_conntrack_htable_size, 0600);
	334
32292a7f	335	MODULE_ALIAS("nf_conntrack-" __stringify(AF_INET));
d2483dde	336	MODULE_ALIAS("ip_conntrack");
32292a7f PM	337	MODULE_LICENSE("GPL");
32292a7f PM	338
0e54d217 DC	339	static struct nf_conntrack_l4proto *builtin_l4proto4[] = {
	340	&nf_conntrack_l4proto_tcp4,
	341	&nf_conntrack_l4proto_udp4,
	342	&nf_conntrack_l4proto_icmp,
c51d3901 DC	343	#ifdef CONFIG_NF_CT_PROTO_DCCP
	344	&nf_conntrack_l4proto_dccp4,
	345	#endif
a85406af DC	346	#ifdef CONFIG_NF_CT_PROTO_SCTP
	347	&nf_conntrack_l4proto_sctp4,
	348	#endif
0e54d217 DC	349	};
0e54d217 DC	350
3ea04dd3 G	351	static int ipv4_net_init(struct net *net)
	352	{
	353	int ret = 0;
	354
0e54d217 DC	355	ret = nf_ct_l4proto_pernet_register(net, builtin_l4proto4,
	356	ARRAY_SIZE(builtin_l4proto4));
	357	if (ret < 0)
	358	return ret;
6330750d	359	ret = nf_ct_l3proto_pernet_register(net, &nf_conntrack_l3proto_ipv4);
3ea04dd3	360	if (ret < 0) {
6330750d	361	pr_err("nf_conntrack_ipv4: pernet registration failed\n");
0e54d217 DC	362	nf_ct_l4proto_pernet_unregister(net, builtin_l4proto4,
0e54d217 DC	363	ARRAY_SIZE(builtin_l4proto4));
3ea04dd3	364	}
3ea04dd3 G	365	return ret;
	366	}
	367
	368	static void ipv4_net_exit(struct net *net)
	369	{
6330750d	370	nf_ct_l3proto_pernet_unregister(net, &nf_conntrack_l3proto_ipv4);
0e54d217 DC	371	nf_ct_l4proto_pernet_unregister(net, builtin_l4proto4,
0e54d217 DC	372	ARRAY_SIZE(builtin_l4proto4));
3ea04dd3 G	373	}
	374
	375	static struct pernet_operations ipv4_net_ops = {
	376	.init = ipv4_net_init,
	377	.exit = ipv4_net_exit,
	378	};
	379
32292a7f	380	static int __init nf_conntrack_l3proto_ipv4_init(void)
9fb9cbb1 YK	381	{
	382	int ret = 0;
	383
32292a7f	384	need_conntrack();
73e4022f	385	nf_defrag_ipv4_enable();
9fb9cbb1 YK	386
	387	ret = nf_register_sockopt(&so_getorigdst);
	388	if (ret < 0) {
09605cc1	389	pr_err("Unable to register netfilter socket option\n");
32292a7f	390	return ret;
9fb9cbb1 YK	391	}
9fb9cbb1 YK	392
3ea04dd3	393	ret = register_pernet_subsys(&ipv4_net_ops);
9fb9cbb1	394	if (ret < 0) {
3ea04dd3	395	pr_err("nf_conntrack_ipv4: can't register pernet ops\n");
9fb9cbb1 YK	396	goto cleanup_sockopt;
	397	}
	398
964ddaa1 PM	399	ret = nf_register_hooks(ipv4_conntrack_ops,
964ddaa1 PM	400	ARRAY_SIZE(ipv4_conntrack_ops));
9fb9cbb1	401	if (ret < 0) {
654d0fbd	402	pr_err("nf_conntrack_ipv4: can't register hooks.\n");
3ea04dd3	403	goto cleanup_pernet;
9fb9cbb1	404	}
6330750d	405
0e54d217 DC	406	ret = nf_ct_l4proto_register(builtin_l4proto4,
	407	ARRAY_SIZE(builtin_l4proto4));
	408	if (ret < 0)
c296bb4d	409	goto cleanup_hooks;
c296bb4d	410
6330750d G	411	ret = nf_ct_l3proto_register(&nf_conntrack_l3proto_ipv4);
	412	if (ret < 0) {
	413	pr_err("nf_conntrack_ipv4: can't register ipv4 proto.\n");
0e54d217	414	goto cleanup_l4proto;
6330750d G	415	}
6330750d G	416
9fb9cbb1	417	return ret;
0e54d217 DC	418	cleanup_l4proto:
	419	nf_ct_l4proto_unregister(builtin_l4proto4,
	420	ARRAY_SIZE(builtin_l4proto4));
e4bd8bce	421	cleanup_hooks:
e905a9ed	422	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
3ea04dd3 G	423	cleanup_pernet:
3ea04dd3 G	424	unregister_pernet_subsys(&ipv4_net_ops);
9fb9cbb1 YK	425	cleanup_sockopt:
9fb9cbb1 YK	426	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	427	return ret;
	428	}
	429
65b4b4e8	430	static void __exit nf_conntrack_l3proto_ipv4_fini(void)
9fb9cbb1	431	{
32292a7f	432	synchronize_net();
6330750d	433	nf_ct_l3proto_unregister(&nf_conntrack_l3proto_ipv4);
0e54d217 DC	434	nf_ct_l4proto_unregister(builtin_l4proto4,
0e54d217 DC	435	ARRAY_SIZE(builtin_l4proto4));
32292a7f	436	nf_unregister_hooks(ipv4_conntrack_ops, ARRAY_SIZE(ipv4_conntrack_ops));
3ea04dd3	437	unregister_pernet_subsys(&ipv4_net_ops);
32292a7f	438	nf_unregister_sockopt(&so_getorigdst);
9fb9cbb1 YK	439	}
9fb9cbb1 YK	440
65b4b4e8 AM	441	module_init(nf_conntrack_l3proto_ipv4_init);
65b4b4e8 AM	442	module_exit(nf_conntrack_l3proto_ipv4_fini);