[deliverable/linux.git] / net / ipv4 / netfilter / ip_nat_standalone.c

/* This file contains all the functions required for the standalone
   ip_nat module.

   These are not required by the compatibility layer.
*/

/* (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.
 */

/*
 * 23 Apr 2001: Harald Welte <laforge@gnumonks.org>
 * 	- new API and handling of conntrack/nat helpers
 * 	- now capable of multiple expectations for one master
 * */

#include <linux/config.h>
#include <linux/types.h>
#include <linux/icmp.h>
#include <linux/ip.h>
#include <linux/netfilter.h>
#include <linux/netfilter_ipv4.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <net/ip.h>
#include <net/checksum.h>
#include <linux/spinlock.h>

#define ASSERT_READ_LOCK(x)
#define ASSERT_WRITE_LOCK(x)

#include <linux/netfilter_ipv4/ip_nat.h>
#include <linux/netfilter_ipv4/ip_nat_rule.h>
#include <linux/netfilter_ipv4/ip_nat_protocol.h>
#include <linux/netfilter_ipv4/ip_nat_core.h>
#include <linux/netfilter_ipv4/ip_nat_helper.h>
#include <linux/netfilter_ipv4/ip_tables.h>
#include <linux/netfilter_ipv4/ip_conntrack_core.h>
#include <linux/netfilter_ipv4/listhelp.h>

#if 0
#define DEBUGP printk
#else
#define DEBUGP(format, args...)
#endif

#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING"  \
			   : ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \
			      : ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT"  \
			         : ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN"  \
				    : "*ERROR*")))

#ifdef CONFIG_XFRM
static void nat_decode_session(struct sk_buff *skb, struct flowi *fl)
{
	struct ip_conntrack *ct;
	struct ip_conntrack_tuple *t;
	enum ip_conntrack_info ctinfo;
	enum ip_conntrack_dir dir;
	unsigned long statusbit;

	ct = ip_conntrack_get(skb, &ctinfo);
	if (ct == NULL)
		return;
	dir = CTINFO2DIR(ctinfo);
	t = &ct->tuplehash[dir].tuple;

	if (dir == IP_CT_DIR_ORIGINAL)
		statusbit = IPS_DST_NAT;
	else
		statusbit = IPS_SRC_NAT;

	if (ct->status & statusbit) {
		fl->fl4_dst = t->dst.ip;
		if (t->dst.protonum == IPPROTO_TCP ||
		    t->dst.protonum == IPPROTO_UDP)
			fl->fl_ip_dport = t->dst.u.tcp.port;
	}

	statusbit ^= IPS_NAT_MASK;

	if (ct->status & statusbit) {
		fl->fl4_src = t->src.ip;
		if (t->dst.protonum == IPPROTO_TCP ||
		    t->dst.protonum == IPPROTO_UDP)
			fl->fl_ip_sport = t->src.u.tcp.port;
	}
}
#endif
		
static unsigned int
ip_nat_fn(unsigned int hooknum,
	  struct sk_buff **pskb,
	  const struct net_device *in,
	  const struct net_device *out,
	  int (*okfn)(struct sk_buff *))
{
	struct ip_conntrack *ct;
	enum ip_conntrack_info ctinfo;
	struct ip_nat_info *info;
	/* maniptype == SRC for postrouting. */
	enum ip_nat_manip_type maniptype = HOOK2MANIP(hooknum);

	/* We never see fragments: conntrack defrags on pre-routing
	   and local-out, and ip_nat_out protects post-routing. */
	IP_NF_ASSERT(!((*pskb)->nh.iph->frag_off
		       & htons(IP_MF|IP_OFFSET)));

	/* If we had a hardware checksum before, it's now invalid */
	if ((*pskb)->ip_summed == CHECKSUM_HW)
		if (skb_checksum_help(*pskb, (out == NULL)))
			return NF_DROP;

	ct = ip_conntrack_get(*pskb, &ctinfo);
	/* Can't track?  It's not due to stress, or conntrack would
	   have dropped it.  Hence it's the user's responsibilty to
	   packet filter it out, or implement conntrack/NAT for that
	   protocol. 8) --RR */
	if (!ct) {
		/* Exception: ICMP redirect to new connection (not in
                   hash table yet).  We must not let this through, in
                   case we're doing NAT to the same network. */
		if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) {
			struct icmphdr _hdr, *hp;

			hp = skb_header_pointer(*pskb,
						(*pskb)->nh.iph->ihl*4,
						sizeof(_hdr), &_hdr);
			if (hp != NULL &&
			    hp->type == ICMP_REDIRECT)
				return NF_DROP;
		}
		return NF_ACCEPT;
	}

	/* Don't try to NAT if this packet is not conntracked */
	if (ct == &ip_conntrack_untracked)
		return NF_ACCEPT;

	switch (ctinfo) {
	case IP_CT_RELATED:
	case IP_CT_RELATED+IP_CT_IS_REPLY:
		if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) {
			if (!ip_nat_icmp_reply_translation(pskb, ct, maniptype,
							   CTINFO2DIR(ctinfo)))
				return NF_DROP;
			else
				return NF_ACCEPT;
		}
		/* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */
	case IP_CT_NEW:
		info = &ct->nat.info;

		/* Seen it before?  This can happen for loopback, retrans,
		   or local packets.. */
		if (!ip_nat_initialized(ct, maniptype)) {
			unsigned int ret;

			if (unlikely(is_confirmed(ct)))
				/* NAT module was loaded late */
				ret = alloc_null_binding_confirmed(ct, info,
				                                   hooknum);
			else if (hooknum == NF_IP_LOCAL_IN)
				/* LOCAL_IN hook doesn't have a chain!  */
				ret = alloc_null_binding(ct, info, hooknum);
			else
				ret = ip_nat_rule_find(pskb, hooknum,
						       in, out, ct,
						       info);

			if (ret != NF_ACCEPT) {
				return ret;
			}
		} else
			DEBUGP("Already setup manip %s for ct %p\n",
			       maniptype == IP_NAT_MANIP_SRC ? "SRC" : "DST",
			       ct);
		break;

	default:
		/* ESTABLISHED */
		IP_NF_ASSERT(ctinfo == IP_CT_ESTABLISHED
			     || ctinfo == (IP_CT_ESTABLISHED+IP_CT_IS_REPLY));
		info = &ct->nat.info;
	}

	IP_NF_ASSERT(info);
	return ip_nat_packet(ct, ctinfo, hooknum, pskb);
}

static unsigned int
ip_nat_in(unsigned int hooknum,
          struct sk_buff **pskb,
          const struct net_device *in,
          const struct net_device *out,
          int (*okfn)(struct sk_buff *))
{
	unsigned int ret;
	u_int32_t daddr = (*pskb)->nh.iph->daddr;

	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
	if (ret != NF_DROP && ret != NF_STOLEN
	    && daddr != (*pskb)->nh.iph->daddr) {
		dst_release((*pskb)->dst);
		(*pskb)->dst = NULL;
	}
	return ret;
}

static unsigned int
ip_nat_out(unsigned int hooknum,
	   struct sk_buff **pskb,
	   const struct net_device *in,
	   const struct net_device *out,
	   int (*okfn)(struct sk_buff *))
{
	struct ip_conntrack *ct;
	enum ip_conntrack_info ctinfo;
	unsigned int ret;

	/* root is playing with raw sockets. */
	if ((*pskb)->len < sizeof(struct iphdr)
	    || (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr))
		return NF_ACCEPT;

	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
#ifdef CONFIG_XFRM
	if (ret != NF_DROP && ret != NF_STOLEN
	    && (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);

		if (ct->tuplehash[dir].tuple.src.ip !=
		    ct->tuplehash[!dir].tuple.dst.ip
		    || ct->tuplehash[dir].tuple.src.u.all !=
		       ct->tuplehash[!dir].tuple.dst.u.all
		    )
			return ip_xfrm_me_harder(pskb) == 0 ? ret : NF_DROP;
	}
#endif
	return ret;
}

static unsigned int
ip_nat_local_fn(unsigned int hooknum,
		struct sk_buff **pskb,
		const struct net_device *in,
		const struct net_device *out,
		int (*okfn)(struct sk_buff *))
{
	struct ip_conntrack *ct;
	enum ip_conntrack_info ctinfo;
	unsigned int ret;

	/* root is playing with raw sockets. */
	if ((*pskb)->len < sizeof(struct iphdr)
	    || (*pskb)->nh.iph->ihl * 4 < sizeof(struct iphdr))
		return NF_ACCEPT;

	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
	if (ret != NF_DROP && ret != NF_STOLEN
	    && (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) {
		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);

		if (ct->tuplehash[dir].tuple.dst.ip !=
		    ct->tuplehash[!dir].tuple.src.ip
#ifdef CONFIG_XFRM
		    || ct->tuplehash[dir].tuple.dst.u.all !=
		       ct->tuplehash[!dir].tuple.src.u.all
#endif
		    )
			return ip_route_me_harder(pskb) == 0 ? ret : NF_DROP;
	}
	return ret;
}

static unsigned int
ip_nat_adjust(unsigned int hooknum,
	      struct sk_buff **pskb,
	      const struct net_device *in,
	      const struct net_device *out,
	      int (*okfn)(struct sk_buff *))
{
	struct ip_conntrack *ct;
	enum ip_conntrack_info ctinfo;

	ct = ip_conntrack_get(*pskb, &ctinfo);
	if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
	        DEBUGP("ip_nat_standalone: adjusting sequence number\n");
	        if (!ip_nat_seq_adjust(pskb, ct, ctinfo))
	                return NF_DROP;
	}
	return NF_ACCEPT;
}

/* We must be after connection tracking and before packet filtering. */

/* Before packet filtering, change destination */
static struct nf_hook_ops ip_nat_in_ops = {
	.hook		= ip_nat_in,
	.owner		= THIS_MODULE,
	.pf		= PF_INET,
	.hooknum	= NF_IP_PRE_ROUTING,
	.priority	= NF_IP_PRI_NAT_DST,
};

/* After packet filtering, change source */
static struct nf_hook_ops ip_nat_out_ops = {
	.hook		= ip_nat_out,
	.owner		= THIS_MODULE,
	.pf		= PF_INET,
	.hooknum	= NF_IP_POST_ROUTING,
	.priority	= NF_IP_PRI_NAT_SRC,
};

/* After conntrack, adjust sequence number */
static struct nf_hook_ops ip_nat_adjust_out_ops = {
	.hook		= ip_nat_adjust,
	.owner		= THIS_MODULE,
	.pf		= PF_INET,
	.hooknum	= NF_IP_POST_ROUTING,
	.priority	= NF_IP_PRI_NAT_SEQ_ADJUST,
};

/* Before packet filtering, change destination */
static struct nf_hook_ops ip_nat_local_out_ops = {
	.hook		= ip_nat_local_fn,
	.owner		= THIS_MODULE,
	.pf		= PF_INET,
	.hooknum	= NF_IP_LOCAL_OUT,
	.priority	= NF_IP_PRI_NAT_DST,
};

/* After packet filtering, change source for reply packets of LOCAL_OUT DNAT */
static struct nf_hook_ops ip_nat_local_in_ops = {
	.hook		= ip_nat_fn,
	.owner		= THIS_MODULE,
	.pf		= PF_INET,
	.hooknum	= NF_IP_LOCAL_IN,
	.priority	= NF_IP_PRI_NAT_SRC,
};

/* After conntrack, adjust sequence number */
static struct nf_hook_ops ip_nat_adjust_in_ops = {
	.hook		= ip_nat_adjust,
	.owner		= THIS_MODULE,
	.pf		= PF_INET,
	.hooknum	= NF_IP_LOCAL_IN,
	.priority	= NF_IP_PRI_NAT_SEQ_ADJUST,
};


static int init_or_cleanup(int init)
{
	int ret = 0;

	need_conntrack();

	if (!init) goto cleanup;

#ifdef CONFIG_XFRM
	BUG_ON(ip_nat_decode_session != NULL);
	ip_nat_decode_session = nat_decode_session;
#endif
	ret = ip_nat_rule_init();
	if (ret < 0) {
		printk("ip_nat_init: can't setup rules.\n");
		goto cleanup_decode_session;
	}
	ret = nf_register_hook(&ip_nat_in_ops);
	if (ret < 0) {
		printk("ip_nat_init: can't register in hook.\n");
		goto cleanup_rule_init;
	}
	ret = nf_register_hook(&ip_nat_out_ops);
	if (ret < 0) {
		printk("ip_nat_init: can't register out hook.\n");
		goto cleanup_inops;
	}
	ret = nf_register_hook(&ip_nat_adjust_in_ops);
	if (ret < 0) {
		printk("ip_nat_init: can't register adjust in hook.\n");
		goto cleanup_outops;
	}
	ret = nf_register_hook(&ip_nat_adjust_out_ops);
	if (ret < 0) {
		printk("ip_nat_init: can't register adjust out hook.\n");
		goto cleanup_adjustin_ops;
	}
	ret = nf_register_hook(&ip_nat_local_out_ops);
	if (ret < 0) {
		printk("ip_nat_init: can't register local out hook.\n");
		goto cleanup_adjustout_ops;
	}
	ret = nf_register_hook(&ip_nat_local_in_ops);
	if (ret < 0) {
		printk("ip_nat_init: can't register local in hook.\n");
		goto cleanup_localoutops;
	}
	return ret;

 cleanup:
	nf_unregister_hook(&ip_nat_local_in_ops);
 cleanup_localoutops:
	nf_unregister_hook(&ip_nat_local_out_ops);
 cleanup_adjustout_ops:
	nf_unregister_hook(&ip_nat_adjust_out_ops);
 cleanup_adjustin_ops:
	nf_unregister_hook(&ip_nat_adjust_in_ops);
 cleanup_outops:
	nf_unregister_hook(&ip_nat_out_ops);
 cleanup_inops:
	nf_unregister_hook(&ip_nat_in_ops);
 cleanup_rule_init:
	ip_nat_rule_cleanup();
 cleanup_decode_session:
#ifdef CONFIG_XFRM
	ip_nat_decode_session = NULL;
	synchronize_net();
#endif
	return ret;
}

static int __init init(void)
{
	return init_or_cleanup(1);
}

static void __exit fini(void)
{
	init_or_cleanup(0);
}

module_init(init);
module_exit(fini);

MODULE_LICENSE("GPL");
Commit	Line	Data
1da177e4 LT	1	/* This file contains all the functions required for the standalone
	2	ip_nat module.
	3
	4	These are not required by the compatibility layer.
	5	*/
	6
	7	/* (C) 1999-2001 Paul `Rusty' Russell
	8	* (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org>
	9	*
	10	* This program is free software; you can redistribute it and/or modify
	11	* it under the terms of the GNU General Public License version 2 as
	12	* published by the Free Software Foundation.
	13	*/
	14
	15	/*
	16	* 23 Apr 2001: Harald Welte <laforge@gnumonks.org>
	17	* - new API and handling of conntrack/nat helpers
	18	* - now capable of multiple expectations for one master
	19	* */
	20
	21	#include <linux/config.h>
	22	#include <linux/types.h>
	23	#include <linux/icmp.h>
	24	#include <linux/ip.h>
	25	#include <linux/netfilter.h>
	26	#include <linux/netfilter_ipv4.h>
	27	#include <linux/module.h>
	28	#include <linux/skbuff.h>
	29	#include <linux/proc_fs.h>
	30	#include <net/ip.h>
	31	#include <net/checksum.h>
	32	#include <linux/spinlock.h>
	33
e45b1be8 PM	34	#define ASSERT_READ_LOCK(x)
e45b1be8 PM	35	#define ASSERT_WRITE_LOCK(x)
1da177e4 LT	36
	37	#include <linux/netfilter_ipv4/ip_nat.h>
	38	#include <linux/netfilter_ipv4/ip_nat_rule.h>
	39	#include <linux/netfilter_ipv4/ip_nat_protocol.h>
	40	#include <linux/netfilter_ipv4/ip_nat_core.h>
	41	#include <linux/netfilter_ipv4/ip_nat_helper.h>
	42	#include <linux/netfilter_ipv4/ip_tables.h>
	43	#include <linux/netfilter_ipv4/ip_conntrack_core.h>
	44	#include <linux/netfilter_ipv4/listhelp.h>
	45
	46	#if 0
	47	#define DEBUGP printk
	48	#else
	49	#define DEBUGP(format, args...)
	50	#endif
	51
	52	#define HOOKNAME(hooknum) ((hooknum) == NF_IP_POST_ROUTING ? "POST_ROUTING" \
	53	: ((hooknum) == NF_IP_PRE_ROUTING ? "PRE_ROUTING" \
	54	: ((hooknum) == NF_IP_LOCAL_OUT ? "LOCAL_OUT" \
	55	: ((hooknum) == NF_IP_LOCAL_IN ? "LOCAL_IN" \
	56	: "ERROR")))
	57
eb9c7ebe PM	58	#ifdef CONFIG_XFRM
	59	static void nat_decode_session(struct sk_buff skb, struct flowi fl)
	60	{
	61	struct ip_conntrack *ct;
	62	struct ip_conntrack_tuple *t;
	63	enum ip_conntrack_info ctinfo;
	64	enum ip_conntrack_dir dir;
	65	unsigned long statusbit;
	66
	67	ct = ip_conntrack_get(skb, &ctinfo);
	68	if (ct == NULL)
	69	return;
	70	dir = CTINFO2DIR(ctinfo);
	71	t = &ct->tuplehash[dir].tuple;
	72
	73	if (dir == IP_CT_DIR_ORIGINAL)
	74	statusbit = IPS_DST_NAT;
	75	else
	76	statusbit = IPS_SRC_NAT;
	77
	78	if (ct->status & statusbit) {
	79	fl->fl4_dst = t->dst.ip;
	80	if (t->dst.protonum == IPPROTO_TCP \|\|
	81	t->dst.protonum == IPPROTO_UDP)
	82	fl->fl_ip_dport = t->dst.u.tcp.port;
	83	}
	84
	85	statusbit ^= IPS_NAT_MASK;
	86
	87	if (ct->status & statusbit) {
	88	fl->fl4_src = t->src.ip;
	89	if (t->dst.protonum == IPPROTO_TCP \|\|
	90	t->dst.protonum == IPPROTO_UDP)
	91	fl->fl_ip_sport = t->src.u.tcp.port;
	92	}
	93	}
	94	#endif
	95
1da177e4 LT	96	static unsigned int
	97	ip_nat_fn(unsigned int hooknum,
	98	struct sk_buff **pskb,
	99	const struct net_device *in,
	100	const struct net_device *out,
	101	int (okfn)(struct sk_buff ))
	102	{
	103	struct ip_conntrack *ct;
	104	enum ip_conntrack_info ctinfo;
	105	struct ip_nat_info *info;
	106	/* maniptype == SRC for postrouting. */
	107	enum ip_nat_manip_type maniptype = HOOK2MANIP(hooknum);
	108
	109	/* We never see fragments: conntrack defrags on pre-routing
	110	and local-out, and ip_nat_out protects post-routing. */
	111	IP_NF_ASSERT(!((*pskb)->nh.iph->frag_off
	112	& htons(IP_MF\|IP_OFFSET)));
	113
1da177e4 LT	114	/* If we had a hardware checksum before, it's now invalid */
	115	if ((*pskb)->ip_summed == CHECKSUM_HW)
	116	if (skb_checksum_help(*pskb, (out == NULL)))
	117	return NF_DROP;
	118
	119	ct = ip_conntrack_get(*pskb, &ctinfo);
	120	/* Can't track? It's not due to stress, or conntrack would
	121	have dropped it. Hence it's the user's responsibilty to
	122	packet filter it out, or implement conntrack/NAT for that
	123	protocol. 8) --RR */
	124	if (!ct) {
	125	/* Exception: ICMP redirect to new connection (not in
	126	hash table yet). We must not let this through, in
	127	case we're doing NAT to the same network. */
	128	if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) {
	129	struct icmphdr _hdr, *hp;
	130
	131	hp = skb_header_pointer(*pskb,
	132	(pskb)->nh.iph->ihl4,
	133	sizeof(_hdr), &_hdr);
	134	if (hp != NULL &&
	135	hp->type == ICMP_REDIRECT)
	136	return NF_DROP;
	137	}
	138	return NF_ACCEPT;
	139	}
	140
8b83bc77 HW	141	/* Don't try to NAT if this packet is not conntracked */
	142	if (ct == &ip_conntrack_untracked)
	143	return NF_ACCEPT;
	144
1da177e4 LT	145	switch (ctinfo) {
	146	case IP_CT_RELATED:
	147	case IP_CT_RELATED+IP_CT_IS_REPLY:
	148	if ((*pskb)->nh.iph->protocol == IPPROTO_ICMP) {
188bab3a HW	149	if (!ip_nat_icmp_reply_translation(pskb, ct, maniptype,
188bab3a HW	150	CTINFO2DIR(ctinfo)))
1da177e4 LT	151	return NF_DROP;
	152	else
	153	return NF_ACCEPT;
	154	}
	155	/* Fall thru... (Only ICMPs can be IP_CT_IS_REPLY) */
	156	case IP_CT_NEW:
	157	info = &ct->nat.info;
	158
	159	/* Seen it before? This can happen for loopback, retrans,
	160	or local packets.. */
	161	if (!ip_nat_initialized(ct, maniptype)) {
	162	unsigned int ret;
	163
03486a4f PM	164	if (unlikely(is_confirmed(ct)))
	165	/* NAT module was loaded late */
	166	ret = alloc_null_binding_confirmed(ct, info,
	167	hooknum);
	168	else if (hooknum == NF_IP_LOCAL_IN)
	169	/* LOCAL_IN hook doesn't have a chain! */
1da177e4 LT	170	ret = alloc_null_binding(ct, info, hooknum);
	171	else
	172	ret = ip_nat_rule_find(pskb, hooknum,
	173	in, out, ct,
	174	info);
	175
	176	if (ret != NF_ACCEPT) {
	177	return ret;
	178	}
	179	} else
	180	DEBUGP("Already setup manip %s for ct %p\n",
	181	maniptype == IP_NAT_MANIP_SRC ? "SRC" : "DST",
	182	ct);
	183	break;
	184
	185	default:
	186	/* ESTABLISHED */
	187	IP_NF_ASSERT(ctinfo == IP_CT_ESTABLISHED
	188	\|\| ctinfo == (IP_CT_ESTABLISHED+IP_CT_IS_REPLY));
	189	info = &ct->nat.info;
	190	}
	191
	192	IP_NF_ASSERT(info);
188bab3a	193	return ip_nat_packet(ct, ctinfo, hooknum, pskb);
1da177e4 LT	194	}
	195
	196	static unsigned int
	197	ip_nat_in(unsigned int hooknum,
	198	struct sk_buff **pskb,
	199	const struct net_device *in,
	200	const struct net_device *out,
	201	int (okfn)(struct sk_buff ))
	202	{
1da177e4	203	unsigned int ret;
8e249f08	204	u_int32_t daddr = (*pskb)->nh.iph->daddr;
1da177e4	205
1da177e4 LT	206	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
1da177e4 LT	207	if (ret != NF_DROP && ret != NF_STOLEN
8e249f08 PM	208	&& daddr != (*pskb)->nh.iph->daddr) {
	209	dst_release((*pskb)->dst);
	210	(*pskb)->dst = NULL;
1da177e4 LT	211	}
	212	return ret;
	213	}
	214
	215	static unsigned int
	216	ip_nat_out(unsigned int hooknum,
	217	struct sk_buff **pskb,
	218	const struct net_device *in,
	219	const struct net_device *out,
	220	int (okfn)(struct sk_buff ))
	221	{
5c901daa PM	222	struct ip_conntrack *ct;
	223	enum ip_conntrack_info ctinfo;
	224	unsigned int ret;
	225
1da177e4 LT	226	/* root is playing with raw sockets. */
	227	if ((*pskb)->len < sizeof(struct iphdr)
	228	\|\| (pskb)->nh.iph->ihl 4 < sizeof(struct iphdr))
	229	return NF_ACCEPT;
	230
5c901daa	231	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
ee68cea2	232	#ifdef CONFIG_XFRM
5c901daa PM	233	if (ret != NF_DROP && ret != NF_STOLEN
	234	&& (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) {
	235	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
	236
	237	if (ct->tuplehash[dir].tuple.src.ip !=
	238	ct->tuplehash[!dir].tuple.dst.ip
5c901daa PM	239	\|\| ct->tuplehash[dir].tuple.src.u.all !=
5c901daa PM	240	ct->tuplehash[!dir].tuple.dst.u.all
5c901daa	241	)
ee68cea2	242	return ip_xfrm_me_harder(pskb) == 0 ? ret : NF_DROP;
5c901daa	243	}
ee68cea2	244	#endif
5c901daa	245	return ret;
1da177e4 LT	246	}
	247
	248	static unsigned int
	249	ip_nat_local_fn(unsigned int hooknum,
	250	struct sk_buff **pskb,
	251	const struct net_device *in,
	252	const struct net_device *out,
	253	int (okfn)(struct sk_buff ))
	254	{
4e8e9de7 PM	255	struct ip_conntrack *ct;
4e8e9de7 PM	256	enum ip_conntrack_info ctinfo;
1da177e4 LT	257	unsigned int ret;
	258
	259	/* root is playing with raw sockets. */
	260	if ((*pskb)->len < sizeof(struct iphdr)
	261	\|\| (pskb)->nh.iph->ihl 4 < sizeof(struct iphdr))
	262	return NF_ACCEPT;
	263
1da177e4 LT	264	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
1da177e4 LT	265	if (ret != NF_DROP && ret != NF_STOLEN
4e8e9de7 PM	266	&& (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) {
	267	enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
	268
	269	if (ct->tuplehash[dir].tuple.dst.ip !=
5c901daa PM	270	ct->tuplehash[!dir].tuple.src.ip
	271	#ifdef CONFIG_XFRM
	272	\|\| ct->tuplehash[dir].tuple.dst.u.all !=
8e249f08	273	ct->tuplehash[!dir].tuple.src.u.all
5c901daa PM	274	#endif
5c901daa PM	275	)
4e8e9de7 PM	276	return ip_route_me_harder(pskb) == 0 ? ret : NF_DROP;
4e8e9de7 PM	277	}
1da177e4 LT	278	return ret;
	279	}
	280
e281e3ac PM	281	static unsigned int
	282	ip_nat_adjust(unsigned int hooknum,
	283	struct sk_buff **pskb,
	284	const struct net_device *in,
	285	const struct net_device *out,
	286	int (okfn)(struct sk_buff ))
	287	{
	288	struct ip_conntrack *ct;
	289	enum ip_conntrack_info ctinfo;
	290
	291	ct = ip_conntrack_get(*pskb, &ctinfo);
	292	if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) {
	293	DEBUGP("ip_nat_standalone: adjusting sequence number\n");
	294	if (!ip_nat_seq_adjust(pskb, ct, ctinfo))
	295	return NF_DROP;
	296	}
	297	return NF_ACCEPT;
	298	}
	299
1da177e4 LT	300	/* We must be after connection tracking and before packet filtering. */
	301
	302	/* Before packet filtering, change destination */
	303	static struct nf_hook_ops ip_nat_in_ops = {
	304	.hook = ip_nat_in,
	305	.owner = THIS_MODULE,
	306	.pf = PF_INET,
	307	.hooknum = NF_IP_PRE_ROUTING,
	308	.priority = NF_IP_PRI_NAT_DST,
	309	};
	310
	311	/* After packet filtering, change source */
	312	static struct nf_hook_ops ip_nat_out_ops = {
	313	.hook = ip_nat_out,
	314	.owner = THIS_MODULE,
	315	.pf = PF_INET,
	316	.hooknum = NF_IP_POST_ROUTING,
	317	.priority = NF_IP_PRI_NAT_SRC,
	318	};
	319
e281e3ac PM	320	/* After conntrack, adjust sequence number */
	321	static struct nf_hook_ops ip_nat_adjust_out_ops = {
	322	.hook = ip_nat_adjust,
	323	.owner = THIS_MODULE,
	324	.pf = PF_INET,
	325	.hooknum = NF_IP_POST_ROUTING,
	326	.priority = NF_IP_PRI_NAT_SEQ_ADJUST,
	327	};
	328
1da177e4 LT	329	/* Before packet filtering, change destination */
	330	static struct nf_hook_ops ip_nat_local_out_ops = {
	331	.hook = ip_nat_local_fn,
	332	.owner = THIS_MODULE,
	333	.pf = PF_INET,
	334	.hooknum = NF_IP_LOCAL_OUT,
	335	.priority = NF_IP_PRI_NAT_DST,
	336	};
	337
	338	/* After packet filtering, change source for reply packets of LOCAL_OUT DNAT */
	339	static struct nf_hook_ops ip_nat_local_in_ops = {
	340	.hook = ip_nat_fn,
	341	.owner = THIS_MODULE,
	342	.pf = PF_INET,
	343	.hooknum = NF_IP_LOCAL_IN,
	344	.priority = NF_IP_PRI_NAT_SRC,
	345	};
	346
e281e3ac PM	347	/* After conntrack, adjust sequence number */
	348	static struct nf_hook_ops ip_nat_adjust_in_ops = {
	349	.hook = ip_nat_adjust,
	350	.owner = THIS_MODULE,
	351	.pf = PF_INET,
	352	.hooknum = NF_IP_LOCAL_IN,
	353	.priority = NF_IP_PRI_NAT_SEQ_ADJUST,
	354	};
	355
	356
1da177e4 LT	357	static int init_or_cleanup(int init)
	358	{
	359	int ret = 0;
	360
2e4e6a17	361	need_conntrack();
1da177e4 LT	362
	363	if (!init) goto cleanup;
	364
eb9c7ebe PM	365	#ifdef CONFIG_XFRM
	366	BUG_ON(ip_nat_decode_session != NULL);
	367	ip_nat_decode_session = nat_decode_session;
	368	#endif
1da177e4 LT	369	ret = ip_nat_rule_init();
	370	if (ret < 0) {
	371	printk("ip_nat_init: can't setup rules.\n");
eb9c7ebe	372	goto cleanup_decode_session;
1da177e4	373	}
1da177e4 LT	374	ret = nf_register_hook(&ip_nat_in_ops);
	375	if (ret < 0) {
	376	printk("ip_nat_init: can't register in hook.\n");
188bab3a	377	goto cleanup_rule_init;
1da177e4 LT	378	}
	379	ret = nf_register_hook(&ip_nat_out_ops);
	380	if (ret < 0) {
	381	printk("ip_nat_init: can't register out hook.\n");
	382	goto cleanup_inops;
	383	}
e281e3ac PM	384	ret = nf_register_hook(&ip_nat_adjust_in_ops);
	385	if (ret < 0) {
	386	printk("ip_nat_init: can't register adjust in hook.\n");
	387	goto cleanup_outops;
	388	}
	389	ret = nf_register_hook(&ip_nat_adjust_out_ops);
	390	if (ret < 0) {
	391	printk("ip_nat_init: can't register adjust out hook.\n");
	392	goto cleanup_adjustin_ops;
	393	}
1da177e4 LT	394	ret = nf_register_hook(&ip_nat_local_out_ops);
	395	if (ret < 0) {
	396	printk("ip_nat_init: can't register local out hook.\n");
53b3531b	397	goto cleanup_adjustout_ops;
1da177e4 LT	398	}
	399	ret = nf_register_hook(&ip_nat_local_in_ops);
	400	if (ret < 0) {
	401	printk("ip_nat_init: can't register local in hook.\n");
	402	goto cleanup_localoutops;
	403	}
	404	return ret;
	405
	406	cleanup:
	407	nf_unregister_hook(&ip_nat_local_in_ops);
	408	cleanup_localoutops:
	409	nf_unregister_hook(&ip_nat_local_out_ops);
e281e3ac PM	410	cleanup_adjustout_ops:
	411	nf_unregister_hook(&ip_nat_adjust_out_ops);
	412	cleanup_adjustin_ops:
	413	nf_unregister_hook(&ip_nat_adjust_in_ops);
1da177e4 LT	414	cleanup_outops:
	415	nf_unregister_hook(&ip_nat_out_ops);
	416	cleanup_inops:
	417	nf_unregister_hook(&ip_nat_in_ops);
1da177e4 LT	418	cleanup_rule_init:
1da177e4 LT	419	ip_nat_rule_cleanup();
eb9c7ebe PM	420	cleanup_decode_session:
	421	#ifdef CONFIG_XFRM
	422	ip_nat_decode_session = NULL;
	423	synchronize_net();
	424	#endif
1da177e4 LT	425	return ret;
	426	}
	427
	428	static int __init init(void)
	429	{
	430	return init_or_cleanup(1);
	431	}
	432
	433	static void __exit fini(void)
	434	{
	435	init_or_cleanup(0);
	436	}
	437
	438	module_init(init);
	439	module_exit(fini);
	440
1da177e4	441	MODULE_LICENSE("GPL");