[net-next-2.6.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 struct tcf_common *tcf_nat_ht[NAT_TAB_MASK + 1];
static u32 nat_idx_gen;
static DEFINE_RWLOCK(nat_lock);

static struct tcf_hashinfo nat_hash_info = {
	.htab	=	tcf_nat_ht,
	.hmask	=	NAT_TAB_MASK,
	.lock	=	&nat_lock,
};

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 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;
	struct tcf_common *pc;

	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]);

	pc = tcf_hash_check(parm->index, a, bind, &nat_hash_info);
	if (!pc) {
		pc = tcf_hash_create(parm->index, est, a, sizeof(*p), bind,
				     &nat_idx_gen, &nat_hash_info);
		if (IS_ERR(pc))
		    return PTR_ERR(pc);
		p = to_tcf_nat(pc);
		ret = ACT_P_CREATED;
	} else {
		p = to_tcf_nat(pc);
		if (!ovr) {
			tcf_hash_release(pc, bind, &nat_hash_info);
			return -EEXIST;
		}
	}

	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(pc, &nat_hash_info);

	return ret;
}

static int tcf_nat_cleanup(struct tc_action *a, int bind)
{
	struct tcf_nat *p = a->priv;

	return tcf_hash_release(&p->common, bind, &nat_hash_info);
}

static int tcf_nat(struct sk_buff *skb, 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;

	p->tcf_bstats.bytes += qdisc_pkt_len(skb);
	p->tcf_bstats.packets++;

	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;

	NLA_PUT(skb, TCA_NAT_PARMS, sizeof(opt), &opt);
	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);
	NLA_PUT(skb, TCA_NAT_TM, sizeof(t), &t);

	return skb->len;

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

static struct tc_action_ops act_nat_ops = {
	.kind		=	"nat",
	.hinfo		=	&nat_hash_info,
	.type		=	TCA_ACT_NAT,
	.capab		=	TCA_CAP_NONE,
	.owner		=	THIS_MODULE,
	.act		=	tcf_nat,
	.dump		=	tcf_nat_dump,
	.cleanup	=	tcf_nat_cleanup,
	.lookup		=	tcf_hash_search,
	.init		=	tcf_nat_init,
	.walk		=	tcf_generic_walker
};

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

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

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
	33	static struct tcf_common *tcf_nat_ht[NAT_TAB_MASK + 1];
	34	static u32 nat_idx_gen;
	35	static DEFINE_RWLOCK(nat_lock);
	36
	37	static struct tcf_hashinfo nat_hash_info = {
	38	.htab = tcf_nat_ht,
	39	.hmask = NAT_TAB_MASK,
	40	.lock = &nat_lock,
	41	};
	42
53b2bf3f PM	43	static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
	44	[TCA_NAT_PARMS] = { .len = sizeof(struct tc_nat) },
	45	};
	46
7ba699c6	47	static int tcf_nat_init(struct nlattr nla, struct nlattr est,
b4219952 HX	48	struct tc_action *a, int ovr, int bind)
b4219952 HX	49	{
7ba699c6	50	struct nlattr *tb[TCA_NAT_MAX + 1];
b4219952	51	struct tc_nat *parm;
cee63723	52	int ret = 0, err;
b4219952 HX	53	struct tcf_nat *p;
	54	struct tcf_common *pc;
	55
cee63723	56	if (nla == NULL)
b4219952 HX	57	return -EINVAL;
b4219952 HX	58
53b2bf3f	59	err = nla_parse_nested(tb, TCA_NAT_MAX, nla, nat_policy);
cee63723 PM	60	if (err < 0)
	61	return err;
	62
53b2bf3f	63	if (tb[TCA_NAT_PARMS] == NULL)
b4219952	64	return -EINVAL;
7ba699c6	65	parm = nla_data(tb[TCA_NAT_PARMS]);
b4219952 HX	66
	67	pc = tcf_hash_check(parm->index, a, bind, &nat_hash_info);
	68	if (!pc) {
	69	pc = tcf_hash_create(parm->index, est, a, sizeof(*p), bind,
	70	&nat_idx_gen, &nat_hash_info);
0e991ec6 SH	71	if (IS_ERR(pc))
0e991ec6 SH	72	return PTR_ERR(pc);
b4219952 HX	73	p = to_tcf_nat(pc);
	74	ret = ACT_P_CREATED;
	75	} else {
	76	p = to_tcf_nat(pc);
	77	if (!ovr) {
	78	tcf_hash_release(pc, bind, &nat_hash_info);
	79	return -EEXIST;
	80	}
	81	}
	82
	83	spin_lock_bh(&p->tcf_lock);
	84	p->old_addr = parm->old_addr;
	85	p->new_addr = parm->new_addr;
	86	p->mask = parm->mask;
	87	p->flags = parm->flags;
	88
	89	p->tcf_action = parm->action;
	90	spin_unlock_bh(&p->tcf_lock);
	91
	92	if (ret == ACT_P_CREATED)
	93	tcf_hash_insert(pc, &nat_hash_info);
	94
	95	return ret;
	96	}
	97
	98	static int tcf_nat_cleanup(struct tc_action *a, int bind)
	99	{
	100	struct tcf_nat *p = a->priv;
	101
	102	return tcf_hash_release(&p->common, bind, &nat_hash_info);
	103	}
	104
	105	static int tcf_nat(struct sk_buff skb, struct tc_action a,
	106	struct tcf_result *res)
	107	{
	108	struct tcf_nat *p = a->priv;
	109	struct iphdr *iph;
	110	__be32 old_addr;
	111	__be32 new_addr;
	112	__be32 mask;
	113	__be32 addr;
	114	int egress;
	115	int action;
	116	int ihl;
36d12690	117	int noff;
b4219952 HX	118
	119	spin_lock(&p->tcf_lock);
	120
	121	p->tcf_tm.lastuse = jiffies;
	122	old_addr = p->old_addr;
	123	new_addr = p->new_addr;
	124	mask = p->mask;
	125	egress = p->flags & TCA_NAT_FLAG_EGRESS;
	126	action = p->tcf_action;
	127
0abf77e5	128	p->tcf_bstats.bytes += qdisc_pkt_len(skb);
b4219952 HX	129	p->tcf_bstats.packets++;
	130
	131	spin_unlock(&p->tcf_lock);
	132
	133	if (unlikely(action == TC_ACT_SHOT))
	134	goto drop;
	135
36d12690 CG	136	noff = skb_network_offset(skb);
36d12690 CG	137	if (!pskb_may_pull(skb, sizeof(*iph) + noff))
b4219952 HX	138	goto drop;
	139
	140	iph = ip_hdr(skb);
	141
	142	if (egress)
	143	addr = iph->saddr;
	144	else
	145	addr = iph->daddr;
	146
	147	if (!((old_addr ^ addr) & mask)) {
	148	if (skb_cloned(skb) &&
36d12690	149	!skb_clone_writable(skb, sizeof(*iph) + noff) &&
b4219952 HX	150	pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
	151	goto drop;
	152
	153	new_addr &= mask;
	154	new_addr \|= addr & ~mask;
	155
	156	/* Rewrite IP header */
	157	iph = ip_hdr(skb);
	158	if (egress)
	159	iph->saddr = new_addr;
	160	else
	161	iph->daddr = new_addr;
	162
be0ea7d5	163	csum_replace4(&iph->check, addr, new_addr);
33c29dde CG	164	} else if ((iph->frag_off & htons(IP_OFFSET)) \|\|
	165	iph->protocol != IPPROTO_ICMP) {
	166	goto out;
b4219952 HX	167	}
	168
	169	ihl = iph->ihl * 4;
	170
	171	/* It would be nice to share code with stateful NAT. */
	172	switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) {
	173	case IPPROTO_TCP:
	174	{
	175	struct tcphdr *tcph;
	176
36d12690	177	if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) \|\|
b4219952	178	(skb_cloned(skb) &&
36d12690	179	!skb_clone_writable(skb, ihl + sizeof(*tcph) + noff) &&
b4219952 HX	180	pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
	181	goto drop;
	182
	183	tcph = (void *)(skb_network_header(skb) + ihl);
be0ea7d5	184	inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr, 1);
b4219952 HX	185	break;
	186	}
	187	case IPPROTO_UDP:
	188	{
	189	struct udphdr *udph;
	190
36d12690	191	if (!pskb_may_pull(skb, ihl + sizeof(*udph) + noff) \|\|
b4219952	192	(skb_cloned(skb) &&
36d12690	193	!skb_clone_writable(skb, ihl + sizeof(*udph) + noff) &&
b4219952 HX	194	pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))
	195	goto drop;
	196
	197	udph = (void *)(skb_network_header(skb) + ihl);
	198	if (udph->check \|\| skb->ip_summed == CHECKSUM_PARTIAL) {
be0ea7d5 PM	199	inet_proto_csum_replace4(&udph->check, skb, addr,
be0ea7d5 PM	200	new_addr, 1);
b4219952 HX	201	if (!udph->check)
	202	udph->check = CSUM_MANGLED_0;
	203	}
	204	break;
	205	}
	206	case IPPROTO_ICMP:
	207	{
	208	struct icmphdr *icmph;
	209
36d12690	210	if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + noff))
b4219952 HX	211	goto drop;
	212
	213	icmph = (void *)(skb_network_header(skb) + ihl);
	214
	215	if ((icmph->type != ICMP_DEST_UNREACH) &&
	216	(icmph->type != ICMP_TIME_EXCEEDED) &&
	217	(icmph->type != ICMP_PARAMETERPROB))
	218	break;
	219
36d12690 CG	220	if (!pskb_may_pull(skb, ihl + sizeof(icmph) + sizeof(iph) +
36d12690 CG	221	noff))
70c2efa5 CG	222	goto drop;
70c2efa5 CG	223
072d79a3	224	icmph = (void *)(skb_network_header(skb) + ihl);
b4219952 HX	225	iph = (void *)(icmph + 1);
	226	if (egress)
	227	addr = iph->daddr;
	228	else
	229	addr = iph->saddr;
	230
	231	if ((old_addr ^ addr) & mask)
	232	break;
	233
	234	if (skb_cloned(skb) &&
36d12690 CG	235	!skb_clone_writable(skb, ihl + sizeof(*icmph) +
36d12690 CG	236	sizeof(*iph) + noff) &&
b4219952 HX	237	pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
	238	goto drop;
	239
	240	icmph = (void *)(skb_network_header(skb) + ihl);
	241	iph = (void *)(icmph + 1);
	242
	243	new_addr &= mask;
	244	new_addr \|= addr & ~mask;
	245
	246	/* XXX Fix up the inner checksums. */
	247	if (egress)
	248	iph->daddr = new_addr;
	249	else
	250	iph->saddr = new_addr;
	251
be0ea7d5	252	inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
3a3dfb06	253	0);
b4219952 HX	254	break;
	255	}
	256	default:
	257	break;
	258	}
	259
33c29dde	260	out:
b4219952 HX	261	return action;
	262
	263	drop:
	264	spin_lock(&p->tcf_lock);
	265	p->tcf_qstats.drops++;
	266	spin_unlock(&p->tcf_lock);
	267	return TC_ACT_SHOT;
	268	}
	269
	270	static int tcf_nat_dump(struct sk_buff skb, struct tc_action a,
	271	int bind, int ref)
	272	{
	273	unsigned char *b = skb_tail_pointer(skb);
	274	struct tcf_nat *p = a->priv;
1c40be12 ED	275	struct tc_nat opt = {
	276	.old_addr = p->old_addr,
	277	.new_addr = p->new_addr,
	278	.mask = p->mask,
	279	.flags = p->flags,
	280
	281	.index = p->tcf_index,
	282	.action = p->tcf_action,
	283	.refcnt = p->tcf_refcnt - ref,
	284	.bindcnt = p->tcf_bindcnt - bind,
	285	};
b4219952	286	struct tcf_t t;
b4219952	287
504f85c9	288	NLA_PUT(skb, TCA_NAT_PARMS, sizeof(opt), &opt);
b4219952 HX	289	t.install = jiffies_to_clock_t(jiffies - p->tcf_tm.install);
	290	t.lastuse = jiffies_to_clock_t(jiffies - p->tcf_tm.lastuse);
	291	t.expires = jiffies_to_clock_t(p->tcf_tm.expires);
7ba699c6	292	NLA_PUT(skb, TCA_NAT_TM, sizeof(t), &t);
b4219952	293
b4219952 HX	294	return skb->len;
b4219952 HX	295
7ba699c6	296	nla_put_failure:
b4219952	297	nlmsg_trim(skb, b);
b4219952 HX	298	return -1;
	299	}
	300
	301	static struct tc_action_ops act_nat_ops = {
	302	.kind = "nat",
	303	.hinfo = &nat_hash_info,
	304	.type = TCA_ACT_NAT,
	305	.capab = TCA_CAP_NONE,
	306	.owner = THIS_MODULE,
	307	.act = tcf_nat,
	308	.dump = tcf_nat_dump,
	309	.cleanup = tcf_nat_cleanup,
	310	.lookup = tcf_hash_search,
	311	.init = tcf_nat_init,
	312	.walk = tcf_generic_walker
	313	};
	314
	315	MODULE_DESCRIPTION("Stateless NAT actions");
	316	MODULE_LICENSE("GPL");
	317
	318	static int __init nat_init_module(void)
	319	{
	320	return tcf_register_action(&act_nat_ops);
	321	}
	322
	323	static void __exit nat_cleanup_module(void)
	324	{
	325	tcf_unregister_action(&act_nat_ops);
	326	}
	327
	328	module_init(nat_init_module);
	329	module_exit(nat_cleanup_module);