[deliverable/linux.git] / net / sched / act_nat.c

/*
 * Stateless NAT actions
 *
 * Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License as published by the Free
 * Software Foundation; either version 2 of the License, or (at your option)
 * any later version.
 */

#include <linux/errno.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/netfilter.h>
#include <linux/rtnetlink.h>
#include <linux/skbuff.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/string.h>
#include <linux/tc_act/tc_nat.h>
#include <net/act_api.h>
#include <net/icmp.h>
#include <net/ip.h>
#include <net/netlink.h>
#include <net/tc_act/tc_nat.h>
#include <net/tcp.h>
#include <net/udp.h>


#define NAT_TAB_MASK	15

static int nat_net_id;
static struct tc_action_ops act_nat_ops;

static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
	[TCA_NAT_PARMS]	= { .len = sizeof(struct tc_nat) },
};

static int tcf_nat_init(struct net *net, struct nlattr *nla, struct nlattr *est,
			struct tc_action **a, int ovr, int bind)
{
	struct tc_action_net *tn = net_generic(net, nat_net_id);
	struct nlattr *tb[TCA_NAT_MAX + 1];
	struct tc_nat *parm;
	int ret = 0, err;
	struct tcf_nat *p;

	if (nla == NULL)
		return -EINVAL;

	err = nla_parse_nested(tb, TCA_NAT_MAX, nla, nat_policy);
	if (err < 0)
		return err;

	if (tb[TCA_NAT_PARMS] == NULL)
		return -EINVAL;
	parm = nla_data(tb[TCA_NAT_PARMS]);

	if (!tcf_hash_check(tn, parm->index, a, bind)) {
		ret = tcf_hash_create(tn, parm->index, est, a,
				      &act_nat_ops, bind, false);
		if (ret)
			return ret;
		ret = ACT_P_CREATED;
	} else {
		if (bind)
			return 0;
		tcf_hash_release(*a, bind);
		if (!ovr)
			return -EEXIST;
	}
	p = to_tcf_nat(*a);

	spin_lock_bh(&p->tcf_lock);
	p->old_addr = parm->old_addr;
	p->new_addr = parm->new_addr;
	p->mask = parm->mask;
	p->flags = parm->flags;

	p->tcf_action = parm->action;
	spin_unlock_bh(&p->tcf_lock);

	if (ret == ACT_P_CREATED)
		tcf_hash_insert(tn, *a);

	return ret;
}

static int tcf_nat(struct sk_buff *skb, const struct tc_action *a,
		   struct tcf_result *res)
{
	struct tcf_nat *p = to_tcf_nat(a);
	struct iphdr *iph;
	__be32 old_addr;
	__be32 new_addr;
	__be32 mask;
	__be32 addr;
	int egress;
	int action;
	int ihl;
	int noff;

	spin_lock(&p->tcf_lock);

	tcf_lastuse_update(&p->tcf_tm);
	old_addr = p->old_addr;
	new_addr = p->new_addr;
	mask = p->mask;
	egress = p->flags & TCA_NAT_FLAG_EGRESS;
	action = p->tcf_action;

	bstats_update(&p->tcf_bstats, skb);

	spin_unlock(&p->tcf_lock);

	if (unlikely(action == TC_ACT_SHOT))
		goto drop;

	noff = skb_network_offset(skb);
	if (!pskb_may_pull(skb, sizeof(*iph) + noff))
		goto drop;

	iph = ip_hdr(skb);

	if (egress)
		addr = iph->saddr;
	else
		addr = iph->daddr;

	if (!((old_addr ^ addr) & mask)) {
		if (skb_try_make_writable(skb, sizeof(*iph) + noff))
			goto drop;

		new_addr &= mask;
		new_addr |= addr & ~mask;

		/* Rewrite IP header */
		iph = ip_hdr(skb);
		if (egress)
			iph->saddr = new_addr;
		else
			iph->daddr = new_addr;

		csum_replace4(&iph->check, addr, new_addr);
	} else if ((iph->frag_off & htons(IP_OFFSET)) ||
		   iph->protocol != IPPROTO_ICMP) {
		goto out;
	}

	ihl = iph->ihl * 4;

	/* It would be nice to share code with stateful NAT. */
	switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) {
	case IPPROTO_TCP:
	{
		struct tcphdr *tcph;

		if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) ||
		    skb_try_make_writable(skb, ihl + sizeof(*tcph) + noff))
			goto drop;

		tcph = (void *)(skb_network_header(skb) + ihl);
		inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr,
					 true);
		break;
	}
	case IPPROTO_UDP:
	{
		struct udphdr *udph;

		if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) ||
		    skb_try_make_writable(skb, ihl + sizeof(*udph) + noff))
			goto drop;

		udph = (void *)(skb_network_header(skb) + ihl);
		if (udph->check || skb->ip_summed == CHECKSUM_PARTIAL) {
			inet_proto_csum_replace4(&udph->check, skb, addr,
						 new_addr, true);
			if (!udph->check)
				udph->check = CSUM_MANGLED_0;
		}
		break;
	}
	case IPPROTO_ICMP:
	{
		struct icmphdr *icmph;

		if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + noff))
			goto drop;

		icmph = (void *)(skb_network_header(skb) + ihl);

		if ((icmph->type != ICMP_DEST_UNREACH) &&
		    (icmph->type != ICMP_TIME_EXCEEDED) &&
		    (icmph->type != ICMP_PARAMETERPROB))
			break;

		if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + sizeof(*iph) +
					noff))
			goto drop;

		icmph = (void *)(skb_network_header(skb) + ihl);
		iph = (void *)(icmph + 1);
		if (egress)
			addr = iph->daddr;
		else
			addr = iph->saddr;

		if ((old_addr ^ addr) & mask)
			break;

		if (skb_try_make_writable(skb, ihl + sizeof(*icmph) +
					  sizeof(*iph) + noff))
			goto drop;

		icmph = (void *)(skb_network_header(skb) + ihl);
		iph = (void *)(icmph + 1);

		new_addr &= mask;
		new_addr |= addr & ~mask;

		/* XXX Fix up the inner checksums. */
		if (egress)
			iph->daddr = new_addr;
		else
			iph->saddr = new_addr;

		inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
					 false);
		break;
	}
	default:
		break;
	}

out:
	return action;

drop:
	spin_lock(&p->tcf_lock);
	p->tcf_qstats.drops++;
	spin_unlock(&p->tcf_lock);
	return TC_ACT_SHOT;
}

static int tcf_nat_dump(struct sk_buff *skb, struct tc_action *a,
			int bind, int ref)
{
	unsigned char *b = skb_tail_pointer(skb);
	struct tcf_nat *p = to_tcf_nat(a);
	struct tc_nat opt = {
		.old_addr = p->old_addr,
		.new_addr = p->new_addr,
		.mask     = p->mask,
		.flags    = p->flags,

		.index    = p->tcf_index,
		.action   = p->tcf_action,
		.refcnt   = p->tcf_refcnt - ref,
		.bindcnt  = p->tcf_bindcnt - bind,
	};
	struct tcf_t t;

	if (nla_put(skb, TCA_NAT_PARMS, sizeof(opt), &opt))
		goto nla_put_failure;

	tcf_tm_dump(&t, &p->tcf_tm);
	if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD))
		goto nla_put_failure;

	return skb->len;

nla_put_failure:
	nlmsg_trim(skb, b);
	return -1;
}

static int tcf_nat_walker(struct net *net, struct sk_buff *skb,
			  struct netlink_callback *cb, int type,
			  const struct tc_action_ops *ops)
{
	struct tc_action_net *tn = net_generic(net, nat_net_id);

	return tcf_generic_walker(tn, skb, cb, type, ops);
}

static int tcf_nat_search(struct net *net, struct tc_action **a, u32 index)
{
	struct tc_action_net *tn = net_generic(net, nat_net_id);

	return tcf_hash_search(tn, a, index);
}

static struct tc_action_ops act_nat_ops = {
	.kind		=	"nat",
	.type		=	TCA_ACT_NAT,
	.owner		=	THIS_MODULE,
	.act		=	tcf_nat,
	.dump		=	tcf_nat_dump,
	.init		=	tcf_nat_init,
	.walk		=	tcf_nat_walker,
	.lookup		=	tcf_nat_search,
	.size		=	sizeof(struct tcf_nat),
};

static __net_init int nat_init_net(struct net *net)
{
	struct tc_action_net *tn = net_generic(net, nat_net_id);

	return tc_action_net_init(tn, &act_nat_ops, NAT_TAB_MASK);
}

static void __net_exit nat_exit_net(struct net *net)
{
	struct tc_action_net *tn = net_generic(net, nat_net_id);

	tc_action_net_exit(tn);
}

static struct pernet_operations nat_net_ops = {
	.init = nat_init_net,
	.exit = nat_exit_net,
	.id   = &nat_net_id,
	.size = sizeof(struct tc_action_net),
};

MODULE_DESCRIPTION("Stateless NAT actions");
MODULE_LICENSE("GPL");

static int __init nat_init_module(void)
{
	return tcf_register_action(&act_nat_ops, &nat_net_ops);
}

static void __exit nat_cleanup_module(void)
{
	tcf_unregister_action(&act_nat_ops, &nat_net_ops);
}

module_init(nat_init_module);
module_exit(nat_cleanup_module);
Commit	Line	Data
b4219952 HX	1	/*
	2	* Stateless NAT actions
	3	*
	4	* Copyright (c) 2007 Herbert Xu <herbert@gondor.apana.org.au>
	5	*
	6	* This program is free software; you can redistribute it and/or modify it
	7	* under the terms of the GNU General Public License as published by the Free
	8	* Software Foundation; either version 2 of the License, or (at your option)
	9	* any later version.
	10	*/
	11
	12	#include <linux/errno.h>
	13	#include <linux/init.h>
	14	#include <linux/kernel.h>
	15	#include <linux/module.h>
	16	#include <linux/netfilter.h>
	17	#include <linux/rtnetlink.h>
	18	#include <linux/skbuff.h>
	19	#include <linux/slab.h>
	20	#include <linux/spinlock.h>
	21	#include <linux/string.h>
	22	#include <linux/tc_act/tc_nat.h>
	23	#include <net/act_api.h>
	24	#include <net/icmp.h>
	25	#include <net/ip.h>
	26	#include <net/netlink.h>
	27	#include <net/tc_act/tc_nat.h>
	28	#include <net/tcp.h>
	29	#include <net/udp.h>
	30
	31
	32	#define NAT_TAB_MASK 15
b4219952	33
ddf97ccd	34	static int nat_net_id;
a85a970a	35	static struct tc_action_ops act_nat_ops;
ddf97ccd	36
53b2bf3f PM	37	static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
	38	[TCA_NAT_PARMS] = { .len = sizeof(struct tc_nat) },
	39	};
	40
c1b52739	41	static int tcf_nat_init(struct net net, struct nlattr nla, struct nlattr *est,
a85a970a	42	struct tc_action **a, int ovr, int bind)
b4219952	43	{
ddf97ccd	44	struct tc_action_net *tn = net_generic(net, nat_net_id);
7ba699c6	45	struct nlattr *tb[TCA_NAT_MAX + 1];
b4219952	46	struct tc_nat *parm;
cee63723	47	int ret = 0, err;
b4219952	48	struct tcf_nat *p;
b4219952	49
cee63723	50	if (nla == NULL)
b4219952 HX	51	return -EINVAL;
b4219952 HX	52
53b2bf3f	53	err = nla_parse_nested(tb, TCA_NAT_MAX, nla, nat_policy);
cee63723 PM	54	if (err < 0)
	55	return err;
	56
53b2bf3f	57	if (tb[TCA_NAT_PARMS] == NULL)
b4219952	58	return -EINVAL;
7ba699c6	59	parm = nla_data(tb[TCA_NAT_PARMS]);
b4219952	60
ddf97ccd WC	61	if (!tcf_hash_check(tn, parm->index, a, bind)) {
ddf97ccd WC	62	ret = tcf_hash_create(tn, parm->index, est, a,
a85a970a	63	&act_nat_ops, bind, false);
86062033 WC	64	if (ret)
86062033 WC	65	return ret;
b4219952 HX	66	ret = ACT_P_CREATED;
b4219952 HX	67	} else {
1a29321e JHS	68	if (bind)
1a29321e JHS	69	return 0;
a85a970a	70	tcf_hash_release(*a, bind);
1a29321e	71	if (!ovr)
b4219952	72	return -EEXIST;
b4219952	73	}
a85a970a	74	p = to_tcf_nat(*a);
b4219952 HX	75
	76	spin_lock_bh(&p->tcf_lock);
	77	p->old_addr = parm->old_addr;
	78	p->new_addr = parm->new_addr;
	79	p->mask = parm->mask;
	80	p->flags = parm->flags;
	81
	82	p->tcf_action = parm->action;
	83	spin_unlock_bh(&p->tcf_lock);
	84
	85	if (ret == ACT_P_CREATED)
a85a970a	86	tcf_hash_insert(tn, *a);
b4219952 HX	87
	88	return ret;
	89	}
	90
dc7f9f6e	91	static int tcf_nat(struct sk_buff skb, const struct tc_action a,
b4219952 HX	92	struct tcf_result *res)
b4219952 HX	93	{
a85a970a	94	struct tcf_nat *p = to_tcf_nat(a);
b4219952 HX	95	struct iphdr *iph;
	96	__be32 old_addr;
	97	__be32 new_addr;
	98	__be32 mask;
	99	__be32 addr;
	100	int egress;
	101	int action;
	102	int ihl;
36d12690	103	int noff;
b4219952 HX	104
	105	spin_lock(&p->tcf_lock);
	106
9c4a4e48	107	tcf_lastuse_update(&p->tcf_tm);
b4219952 HX	108	old_addr = p->old_addr;
	109	new_addr = p->new_addr;
	110	mask = p->mask;
	111	egress = p->flags & TCA_NAT_FLAG_EGRESS;
	112	action = p->tcf_action;
	113
bfe0d029	114	bstats_update(&p->tcf_bstats, skb);
b4219952 HX	115
	116	spin_unlock(&p->tcf_lock);
	117
	118	if (unlikely(action == TC_ACT_SHOT))
	119	goto drop;
	120
36d12690 CG	121	noff = skb_network_offset(skb);
36d12690 CG	122	if (!pskb_may_pull(skb, sizeof(*iph) + noff))
b4219952 HX	123	goto drop;
	124
	125	iph = ip_hdr(skb);
	126
	127	if (egress)
	128	addr = iph->saddr;
	129	else
	130	addr = iph->daddr;
	131
	132	if (!((old_addr ^ addr) & mask)) {
3697649f	133	if (skb_try_make_writable(skb, sizeof(*iph) + noff))
b4219952 HX	134	goto drop;
	135
	136	new_addr &= mask;
	137	new_addr \|= addr & ~mask;
	138
	139	/* Rewrite IP header */
	140	iph = ip_hdr(skb);
	141	if (egress)
	142	iph->saddr = new_addr;
	143	else
	144	iph->daddr = new_addr;
	145
be0ea7d5	146	csum_replace4(&iph->check, addr, new_addr);
33c29dde CG	147	} else if ((iph->frag_off & htons(IP_OFFSET)) \|\|
	148	iph->protocol != IPPROTO_ICMP) {
	149	goto out;
b4219952 HX	150	}
	151
	152	ihl = iph->ihl * 4;
	153
	154	/* It would be nice to share code with stateful NAT. */
	155	switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) {
	156	case IPPROTO_TCP:
	157	{
	158	struct tcphdr *tcph;
	159
36d12690	160	if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) \|\|
3697649f	161	skb_try_make_writable(skb, ihl + sizeof(*tcph) + noff))
b4219952 HX	162	goto drop;
	163
	164	tcph = (void *)(skb_network_header(skb) + ihl);
4b048d6d TH	165	inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr,
4b048d6d TH	166	true);
b4219952 HX	167	break;
	168	}
	169	case IPPROTO_UDP:
	170	{
	171	struct udphdr *udph;
	172
36d12690	173	if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) \|\|
3697649f	174	skb_try_make_writable(skb, ihl + sizeof(*udph) + noff))
b4219952 HX	175	goto drop;
	176
	177	udph = (void *)(skb_network_header(skb) + ihl);
	178	if (udph->check \|\| skb->ip_summed == CHECKSUM_PARTIAL) {
be0ea7d5	179	inet_proto_csum_replace4(&udph->check, skb, addr,
4b048d6d	180	new_addr, true);
b4219952 HX	181	if (!udph->check)
	182	udph->check = CSUM_MANGLED_0;
	183	}
	184	break;
	185	}
	186	case IPPROTO_ICMP:
	187	{
	188	struct icmphdr *icmph;
	189
36d12690	190	if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + noff))
b4219952 HX	191	goto drop;
	192
	193	icmph = (void *)(skb_network_header(skb) + ihl);
	194
	195	if ((icmph->type != ICMP_DEST_UNREACH) &&
	196	(icmph->type != ICMP_TIME_EXCEEDED) &&
	197	(icmph->type != ICMP_PARAMETERPROB))
	198	break;
	199
36d12690 CG	200	if (!pskb_may_pull(skb, ihl + sizeof(icmph) + sizeof(iph) +
36d12690 CG	201	noff))
70c2efa5 CG	202	goto drop;
70c2efa5 CG	203
072d79a3	204	icmph = (void *)(skb_network_header(skb) + ihl);
b4219952 HX	205	iph = (void *)(icmph + 1);
	206	if (egress)
	207	addr = iph->daddr;
	208	else
	209	addr = iph->saddr;
	210
	211	if ((old_addr ^ addr) & mask)
	212	break;
	213
3697649f DB	214	if (skb_try_make_writable(skb, ihl + sizeof(*icmph) +
3697649f DB	215	sizeof(*iph) + noff))
b4219952 HX	216	goto drop;
	217
	218	icmph = (void *)(skb_network_header(skb) + ihl);
	219	iph = (void *)(icmph + 1);
	220
	221	new_addr &= mask;
	222	new_addr \|= addr & ~mask;
	223
	224	/* XXX Fix up the inner checksums. */
	225	if (egress)
	226	iph->daddr = new_addr;
	227	else
	228	iph->saddr = new_addr;
	229
be0ea7d5	230	inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
4b048d6d	231	false);
b4219952 HX	232	break;
	233	}
	234	default:
	235	break;
	236	}
	237
33c29dde	238	out:
b4219952 HX	239	return action;
	240
	241	drop:
	242	spin_lock(&p->tcf_lock);
	243	p->tcf_qstats.drops++;
	244	spin_unlock(&p->tcf_lock);
	245	return TC_ACT_SHOT;
	246	}
	247
	248	static int tcf_nat_dump(struct sk_buff skb, struct tc_action a,
	249	int bind, int ref)
	250	{
	251	unsigned char *b = skb_tail_pointer(skb);
a85a970a	252	struct tcf_nat *p = to_tcf_nat(a);
1c40be12 ED	253	struct tc_nat opt = {
	254	.old_addr = p->old_addr,
	255	.new_addr = p->new_addr,
	256	.mask = p->mask,
	257	.flags = p->flags,
	258
	259	.index = p->tcf_index,
	260	.action = p->tcf_action,
	261	.refcnt = p->tcf_refcnt - ref,
	262	.bindcnt = p->tcf_bindcnt - bind,
	263	};
b4219952	264	struct tcf_t t;
b4219952	265
1b34ec43 DM	266	if (nla_put(skb, TCA_NAT_PARMS, sizeof(opt), &opt))
1b34ec43 DM	267	goto nla_put_failure;
48d8ee16 JHS	268
48d8ee16 JHS	269	tcf_tm_dump(&t, &p->tcf_tm);
9854518e	270	if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD))
1b34ec43	271	goto nla_put_failure;
b4219952	272
b4219952 HX	273	return skb->len;
b4219952 HX	274
7ba699c6	275	nla_put_failure:
b4219952	276	nlmsg_trim(skb, b);
b4219952 HX	277	return -1;
	278	}
	279
ddf97ccd WC	280	static int tcf_nat_walker(struct net net, struct sk_buff skb,
ddf97ccd WC	281	struct netlink_callback *cb, int type,
a85a970a	282	const struct tc_action_ops *ops)
ddf97ccd WC	283	{
	284	struct tc_action_net *tn = net_generic(net, nat_net_id);
	285
a85a970a	286	return tcf_generic_walker(tn, skb, cb, type, ops);
ddf97ccd WC	287	}
ddf97ccd WC	288
a85a970a	289	static int tcf_nat_search(struct net net, struct tc_action *a, u32 index)
ddf97ccd WC	290	{
	291	struct tc_action_net *tn = net_generic(net, nat_net_id);
	292
	293	return tcf_hash_search(tn, a, index);
	294	}
	295
b4219952 HX	296	static struct tc_action_ops act_nat_ops = {
b4219952 HX	297	.kind = "nat",
b4219952	298	.type = TCA_ACT_NAT,
b4219952 HX	299	.owner = THIS_MODULE,
	300	.act = tcf_nat,
	301	.dump = tcf_nat_dump,
b4219952	302	.init = tcf_nat_init,
ddf97ccd WC	303	.walk = tcf_nat_walker,
ddf97ccd WC	304	.lookup = tcf_nat_search,
a85a970a	305	.size = sizeof(struct tcf_nat),
ddf97ccd WC	306	};
	307
	308	static __net_init int nat_init_net(struct net *net)
	309	{
	310	struct tc_action_net *tn = net_generic(net, nat_net_id);
	311
	312	return tc_action_net_init(tn, &act_nat_ops, NAT_TAB_MASK);
	313	}
	314
	315	static void __net_exit nat_exit_net(struct net *net)
	316	{
	317	struct tc_action_net *tn = net_generic(net, nat_net_id);
	318
	319	tc_action_net_exit(tn);
	320	}
	321
	322	static struct pernet_operations nat_net_ops = {
	323	.init = nat_init_net,
	324	.exit = nat_exit_net,
	325	.id = &nat_net_id,
	326	.size = sizeof(struct tc_action_net),
b4219952 HX	327	};
	328
	329	MODULE_DESCRIPTION("Stateless NAT actions");
	330	MODULE_LICENSE("GPL");
	331
	332	static int __init nat_init_module(void)
	333	{
ddf97ccd	334	return tcf_register_action(&act_nat_ops, &nat_net_ops);
b4219952 HX	335	}
	336
	337	static void __exit nat_cleanup_module(void)
	338	{
ddf97ccd	339	tcf_unregister_action(&act_nat_ops, &nat_net_ops);
b4219952 HX	340	}
	341
	342	module_init(nat_init_module);
	343	module_exit(nat_cleanup_module);