[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 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 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(parm->index, a, bind)) {
		ret = tcf_hash_create(parm->index, est, a, sizeof(*p), bind);
		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(a);

	return ret;
}

static int tcf_nat(struct sk_buff *skb, const struct tc_action *a,
		   struct tcf_result *res)
{
	struct tcf_nat *p = a->priv;
	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);

	p->tcf_tm.lastuse = jiffies;
	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_cloned(skb) &&
		    !skb_clone_writable(skb, sizeof(*iph) + noff) &&
		    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
			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_cloned(skb) &&
		     !skb_clone_writable(skb, ihl + sizeof(*tcph) + noff) &&
		     pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
			goto drop;

		tcph = (void *)(skb_network_header(skb) + ihl);
		inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr, 1);
		break;
	}
	case IPPROTO_UDP:
	{
		struct udphdr *udph;

		if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) ||
		    (skb_cloned(skb) &&
		     !skb_clone_writable(skb, ihl + sizeof(*udph) + noff) &&
		     pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
			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, 1);
			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_cloned(skb) &&
		    !skb_clone_writable(skb, ihl + sizeof(*icmph) +
					     sizeof(*iph) + noff) &&
		    pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
			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,
					 0);
		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 = a->priv;
	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;
	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
	if (nla_put(skb, TCA_NAT_TM, sizeof(t), &t))
		goto nla_put_failure;

	return skb->len;

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

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,
};

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

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

static void __exit nat_cleanup_module(void)
{
	tcf_unregister_action(&act_nat_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
53b2bf3f PM	34	static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
	35	[TCA_NAT_PARMS] = { .len = sizeof(struct tc_nat) },
	36	};
	37
c1b52739	38	static int tcf_nat_init(struct net net, struct nlattr nla, struct nlattr *est,
b4219952 HX	39	struct tc_action *a, int ovr, int bind)
b4219952 HX	40	{
7ba699c6	41	struct nlattr *tb[TCA_NAT_MAX + 1];
b4219952	42	struct tc_nat *parm;
cee63723	43	int ret = 0, err;
b4219952	44	struct tcf_nat *p;
b4219952	45
cee63723	46	if (nla == NULL)
b4219952 HX	47	return -EINVAL;
b4219952 HX	48
53b2bf3f	49	err = nla_parse_nested(tb, TCA_NAT_MAX, nla, nat_policy);
cee63723 PM	50	if (err < 0)
	51	return err;
	52
53b2bf3f	53	if (tb[TCA_NAT_PARMS] == NULL)
b4219952	54	return -EINVAL;
7ba699c6	55	parm = nla_data(tb[TCA_NAT_PARMS]);
b4219952	56
86062033 WC	57	if (!tcf_hash_check(parm->index, a, bind)) {
	58	ret = tcf_hash_create(parm->index, est, a, sizeof(*p), bind);
	59	if (ret)
	60	return ret;
b4219952 HX	61	ret = ACT_P_CREATED;
b4219952 HX	62	} else {
1a29321e JHS	63	if (bind)
1a29321e JHS	64	return 0;
86062033	65	tcf_hash_release(a, bind);
1a29321e	66	if (!ovr)
b4219952	67	return -EEXIST;
b4219952	68	}
86062033	69	p = to_tcf_nat(a);
b4219952 HX	70
	71	spin_lock_bh(&p->tcf_lock);
	72	p->old_addr = parm->old_addr;
	73	p->new_addr = parm->new_addr;
	74	p->mask = parm->mask;
	75	p->flags = parm->flags;
	76
	77	p->tcf_action = parm->action;
	78	spin_unlock_bh(&p->tcf_lock);
	79
	80	if (ret == ACT_P_CREATED)
86062033	81	tcf_hash_insert(a);
b4219952 HX	82
	83	return ret;
	84	}
	85
dc7f9f6e	86	static int tcf_nat(struct sk_buff skb, const struct tc_action a,
b4219952 HX	87	struct tcf_result *res)
	88	{
	89	struct tcf_nat *p = a->priv;
	90	struct iphdr *iph;
	91	__be32 old_addr;
	92	__be32 new_addr;
	93	__be32 mask;
	94	__be32 addr;
	95	int egress;
	96	int action;
	97	int ihl;
36d12690	98	int noff;
b4219952 HX	99
	100	spin_lock(&p->tcf_lock);
	101
	102	p->tcf_tm.lastuse = jiffies;
	103	old_addr = p->old_addr;
	104	new_addr = p->new_addr;
	105	mask = p->mask;
	106	egress = p->flags & TCA_NAT_FLAG_EGRESS;
	107	action = p->tcf_action;
	108
bfe0d029	109	bstats_update(&p->tcf_bstats, skb);
b4219952 HX	110
	111	spin_unlock(&p->tcf_lock);
	112
	113	if (unlikely(action == TC_ACT_SHOT))
	114	goto drop;
	115
36d12690 CG	116	noff = skb_network_offset(skb);
36d12690 CG	117	if (!pskb_may_pull(skb, sizeof(*iph) + noff))
b4219952 HX	118	goto drop;
	119
	120	iph = ip_hdr(skb);
	121
	122	if (egress)
	123	addr = iph->saddr;
	124	else
	125	addr = iph->daddr;
	126
	127	if (!((old_addr ^ addr) & mask)) {
	128	if (skb_cloned(skb) &&
36d12690	129	!skb_clone_writable(skb, sizeof(*iph) + noff) &&
b4219952 HX	130	pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
	131	goto drop;
	132
	133	new_addr &= mask;
	134	new_addr \|= addr & ~mask;
	135
	136	/* Rewrite IP header */
	137	iph = ip_hdr(skb);
	138	if (egress)
	139	iph->saddr = new_addr;
	140	else
	141	iph->daddr = new_addr;
	142
be0ea7d5	143	csum_replace4(&iph->check, addr, new_addr);
33c29dde CG	144	} else if ((iph->frag_off & htons(IP_OFFSET)) \|\|
	145	iph->protocol != IPPROTO_ICMP) {
	146	goto out;
b4219952 HX	147	}
	148
	149	ihl = iph->ihl * 4;
	150
	151	/* It would be nice to share code with stateful NAT. */
	152	switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) {
	153	case IPPROTO_TCP:
	154	{
	155	struct tcphdr *tcph;
	156
36d12690	157	if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) \|\|
b4219952	158	(skb_cloned(skb) &&
36d12690	159	!skb_clone_writable(skb, ihl + sizeof(*tcph) + noff) &&
b4219952 HX	160	pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
	161	goto drop;
	162
	163	tcph = (void *)(skb_network_header(skb) + ihl);
be0ea7d5	164	inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr, 1);
b4219952 HX	165	break;
	166	}
	167	case IPPROTO_UDP:
	168	{
	169	struct udphdr *udph;
	170
36d12690	171	if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) \|\|
b4219952	172	(skb_cloned(skb) &&
36d12690	173	!skb_clone_writable(skb, ihl + sizeof(*udph) + noff) &&
b4219952 HX	174	pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
	175	goto drop;
	176
	177	udph = (void *)(skb_network_header(skb) + ihl);
	178	if (udph->check \|\| skb->ip_summed == CHECKSUM_PARTIAL) {
be0ea7d5 PM	179	inet_proto_csum_replace4(&udph->check, skb, addr,
be0ea7d5 PM	180	new_addr, 1);
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
	214	if (skb_cloned(skb) &&
36d12690 CG	215	!skb_clone_writable(skb, ihl + sizeof(*icmph) +
36d12690 CG	216	sizeof(*iph) + noff) &&
b4219952 HX	217	pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
	218	goto drop;
	219
	220	icmph = (void *)(skb_network_header(skb) + ihl);
	221	iph = (void *)(icmph + 1);
	222
	223	new_addr &= mask;
	224	new_addr \|= addr & ~mask;
	225
	226	/* XXX Fix up the inner checksums. */
	227	if (egress)
	228	iph->daddr = new_addr;
	229	else
	230	iph->saddr = new_addr;
	231
be0ea7d5	232	inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
3a3dfb06	233	0);
b4219952 HX	234	break;
	235	}
	236	default:
	237	break;
	238	}
	239
33c29dde	240	out:
b4219952 HX	241	return action;
	242
	243	drop:
	244	spin_lock(&p->tcf_lock);
	245	p->tcf_qstats.drops++;
	246	spin_unlock(&p->tcf_lock);
	247	return TC_ACT_SHOT;
	248	}
	249
	250	static int tcf_nat_dump(struct sk_buff skb, struct tc_action a,
	251	int bind, int ref)
	252	{
	253	unsigned char *b = skb_tail_pointer(skb);
	254	struct tcf_nat *p = a->priv;
1c40be12 ED	255	struct tc_nat opt = {
	256	.old_addr = p->old_addr,
	257	.new_addr = p->new_addr,
	258	.mask = p->mask,
	259	.flags = p->flags,
	260
	261	.index = p->tcf_index,
	262	.action = p->tcf_action,
	263	.refcnt = p->tcf_refcnt - ref,
	264	.bindcnt = p->tcf_bindcnt - bind,
	265	};
b4219952	266	struct tcf_t t;
b4219952	267
1b34ec43 DM	268	if (nla_put(skb, TCA_NAT_PARMS, sizeof(opt), &opt))
1b34ec43 DM	269	goto nla_put_failure;
b4219952 HX	270	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
	271	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
	272	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
1b34ec43 DM	273	if (nla_put(skb, TCA_NAT_TM, sizeof(t), &t))
1b34ec43 DM	274	goto nla_put_failure;
b4219952	275
b4219952 HX	276	return skb->len;
b4219952 HX	277
7ba699c6	278	nla_put_failure:
b4219952	279	nlmsg_trim(skb, b);
b4219952 HX	280	return -1;
	281	}
	282
	283	static struct tc_action_ops act_nat_ops = {
	284	.kind = "nat",
b4219952	285	.type = TCA_ACT_NAT,
b4219952 HX	286	.owner = THIS_MODULE,
	287	.act = tcf_nat,
	288	.dump = tcf_nat_dump,
b4219952	289	.init = tcf_nat_init,
b4219952 HX	290	};
	291
	292	MODULE_DESCRIPTION("Stateless NAT actions");
	293	MODULE_LICENSE("GPL");
	294
	295	static int __init nat_init_module(void)
	296	{
4f1e9d89	297	return tcf_register_action(&act_nat_ops, NAT_TAB_MASK);
b4219952 HX	298	}
	299
	300	static void __exit nat_cleanup_module(void)
	301	{
	302	tcf_unregister_action(&act_nat_ops);
	303	}
	304
	305	module_init(nat_init_module);
	306	module_exit(nat_cleanup_module);