[mirror_ubuntu-bionic-kernel.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>


static unsigned 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, NULL);
	if (err < 0)
		return err;

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

	if (!tcf_idr_check(tn, parm->index, a, bind)) {
		ret = tcf_idr_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_idr_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_idr_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_idr_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);
}

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
c7d03a00	32	static unsigned int nat_net_id;
a85a970a	33	static struct tc_action_ops act_nat_ops;
ddf97ccd	34
53b2bf3f PM	35	static const struct nla_policy nat_policy[TCA_NAT_MAX + 1] = {
	36	[TCA_NAT_PARMS] = { .len = sizeof(struct tc_nat) },
	37	};
	38
c1b52739	39	static int tcf_nat_init(struct net net, struct nlattr nla, struct nlattr *est,
a85a970a	40	struct tc_action **a, int ovr, int bind)
b4219952	41	{
ddf97ccd	42	struct tc_action_net *tn = net_generic(net, nat_net_id);
7ba699c6	43	struct nlattr *tb[TCA_NAT_MAX + 1];
b4219952	44	struct tc_nat *parm;
cee63723	45	int ret = 0, err;
b4219952	46	struct tcf_nat *p;
b4219952	47
cee63723	48	if (nla == NULL)
b4219952 HX	49	return -EINVAL;
b4219952 HX	50
fceb6435	51	err = nla_parse_nested(tb, TCA_NAT_MAX, nla, nat_policy, NULL);
cee63723 PM	52	if (err < 0)
	53	return err;
	54
53b2bf3f	55	if (tb[TCA_NAT_PARMS] == NULL)
b4219952	56	return -EINVAL;
7ba699c6	57	parm = nla_data(tb[TCA_NAT_PARMS]);
b4219952	58
65a206c0 CM	59	if (!tcf_idr_check(tn, parm->index, a, bind)) {
	60	ret = tcf_idr_create(tn, parm->index, est, a,
	61	&act_nat_ops, bind, false);
86062033 WC	62	if (ret)
86062033 WC	63	return ret;
b4219952 HX	64	ret = ACT_P_CREATED;
b4219952 HX	65	} else {
1a29321e JHS	66	if (bind)
1a29321e JHS	67	return 0;
65a206c0	68	tcf_idr_release(*a, bind);
1a29321e	69	if (!ovr)
b4219952	70	return -EEXIST;
b4219952	71	}
a85a970a	72	p = to_tcf_nat(*a);
b4219952 HX	73
	74	spin_lock_bh(&p->tcf_lock);
	75	p->old_addr = parm->old_addr;
	76	p->new_addr = parm->new_addr;
	77	p->mask = parm->mask;
	78	p->flags = parm->flags;
	79
	80	p->tcf_action = parm->action;
	81	spin_unlock_bh(&p->tcf_lock);
	82
	83	if (ret == ACT_P_CREATED)
65a206c0	84	tcf_idr_insert(tn, *a);
b4219952 HX	85
	86	return ret;
	87	}
	88
dc7f9f6e	89	static int tcf_nat(struct sk_buff skb, const struct tc_action a,
b4219952 HX	90	struct tcf_result *res)
b4219952 HX	91	{
a85a970a	92	struct tcf_nat *p = to_tcf_nat(a);
b4219952 HX	93	struct iphdr *iph;
	94	__be32 old_addr;
	95	__be32 new_addr;
	96	__be32 mask;
	97	__be32 addr;
	98	int egress;
	99	int action;
	100	int ihl;
36d12690	101	int noff;
b4219952 HX	102
	103	spin_lock(&p->tcf_lock);
	104
9c4a4e48	105	tcf_lastuse_update(&p->tcf_tm);
b4219952 HX	106	old_addr = p->old_addr;
	107	new_addr = p->new_addr;
	108	mask = p->mask;
	109	egress = p->flags & TCA_NAT_FLAG_EGRESS;
	110	action = p->tcf_action;
	111
bfe0d029	112	bstats_update(&p->tcf_bstats, skb);
b4219952 HX	113
	114	spin_unlock(&p->tcf_lock);
	115
	116	if (unlikely(action == TC_ACT_SHOT))
	117	goto drop;
	118
36d12690 CG	119	noff = skb_network_offset(skb);
36d12690 CG	120	if (!pskb_may_pull(skb, sizeof(*iph) + noff))
b4219952 HX	121	goto drop;
	122
	123	iph = ip_hdr(skb);
	124
	125	if (egress)
	126	addr = iph->saddr;
	127	else
	128	addr = iph->daddr;
	129
	130	if (!((old_addr ^ addr) & mask)) {
3697649f	131	if (skb_try_make_writable(skb, sizeof(*iph) + noff))
b4219952 HX	132	goto drop;
	133
	134	new_addr &= mask;
	135	new_addr \|= addr & ~mask;
	136
	137	/* Rewrite IP header */
	138	iph = ip_hdr(skb);
	139	if (egress)
	140	iph->saddr = new_addr;
	141	else
	142	iph->daddr = new_addr;
	143
be0ea7d5	144	csum_replace4(&iph->check, addr, new_addr);
33c29dde CG	145	} else if ((iph->frag_off & htons(IP_OFFSET)) \|\|
	146	iph->protocol != IPPROTO_ICMP) {
	147	goto out;
b4219952 HX	148	}
	149
	150	ihl = iph->ihl * 4;
	151
	152	/* It would be nice to share code with stateful NAT. */
	153	switch (iph->frag_off & htons(IP_OFFSET) ? 0 : iph->protocol) {
	154	case IPPROTO_TCP:
	155	{
	156	struct tcphdr *tcph;
	157
36d12690	158	if (!pskb_may_pull(skb, ihl + sizeof(*tcph) + noff) \|\|
3697649f	159	skb_try_make_writable(skb, ihl + sizeof(*tcph) + noff))
b4219952 HX	160	goto drop;
	161
	162	tcph = (void *)(skb_network_header(skb) + ihl);
4b048d6d TH	163	inet_proto_csum_replace4(&tcph->check, skb, addr, new_addr,
4b048d6d TH	164	true);
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) \|\|
3697649f	172	skb_try_make_writable(skb, ihl + sizeof(*udph) + noff))
b4219952 HX	173	goto drop;
	174
	175	udph = (void *)(skb_network_header(skb) + ihl);
	176	if (udph->check \|\| skb->ip_summed == CHECKSUM_PARTIAL) {
be0ea7d5	177	inet_proto_csum_replace4(&udph->check, skb, addr,
4b048d6d	178	new_addr, true);
b4219952 HX	179	if (!udph->check)
	180	udph->check = CSUM_MANGLED_0;
	181	}
	182	break;
	183	}
	184	case IPPROTO_ICMP:
	185	{
	186	struct icmphdr *icmph;
	187
36d12690	188	if (!pskb_may_pull(skb, ihl + sizeof(*icmph) + noff))
b4219952 HX	189	goto drop;
	190
	191	icmph = (void *)(skb_network_header(skb) + ihl);
	192
	193	if ((icmph->type != ICMP_DEST_UNREACH) &&
	194	(icmph->type != ICMP_TIME_EXCEEDED) &&
	195	(icmph->type != ICMP_PARAMETERPROB))
	196	break;
	197
36d12690 CG	198	if (!pskb_may_pull(skb, ihl + sizeof(icmph) + sizeof(iph) +
36d12690 CG	199	noff))
70c2efa5 CG	200	goto drop;
70c2efa5 CG	201
072d79a3	202	icmph = (void *)(skb_network_header(skb) + ihl);
b4219952 HX	203	iph = (void *)(icmph + 1);
	204	if (egress)
	205	addr = iph->daddr;
	206	else
	207	addr = iph->saddr;
	208
	209	if ((old_addr ^ addr) & mask)
	210	break;
	211
3697649f DB	212	if (skb_try_make_writable(skb, ihl + sizeof(*icmph) +
3697649f DB	213	sizeof(*iph) + noff))
b4219952 HX	214	goto drop;
	215
	216	icmph = (void *)(skb_network_header(skb) + ihl);
	217	iph = (void *)(icmph + 1);
	218
	219	new_addr &= mask;
	220	new_addr \|= addr & ~mask;
	221
	222	/* XXX Fix up the inner checksums. */
	223	if (egress)
	224	iph->daddr = new_addr;
	225	else
	226	iph->saddr = new_addr;
	227
be0ea7d5	228	inet_proto_csum_replace4(&icmph->checksum, skb, addr, new_addr,
4b048d6d	229	false);
b4219952 HX	230	break;
	231	}
	232	default:
	233	break;
	234	}
	235
33c29dde	236	out:
b4219952 HX	237	return action;
	238
	239	drop:
	240	spin_lock(&p->tcf_lock);
	241	p->tcf_qstats.drops++;
	242	spin_unlock(&p->tcf_lock);
	243	return TC_ACT_SHOT;
	244	}
	245
	246	static int tcf_nat_dump(struct sk_buff skb, struct tc_action a,
	247	int bind, int ref)
	248	{
	249	unsigned char *b = skb_tail_pointer(skb);
a85a970a	250	struct tcf_nat *p = to_tcf_nat(a);
1c40be12 ED	251	struct tc_nat opt = {
	252	.old_addr = p->old_addr,
	253	.new_addr = p->new_addr,
	254	.mask = p->mask,
	255	.flags = p->flags,
	256
	257	.index = p->tcf_index,
	258	.action = p->tcf_action,
	259	.refcnt = p->tcf_refcnt - ref,
	260	.bindcnt = p->tcf_bindcnt - bind,
	261	};
b4219952	262	struct tcf_t t;
b4219952	263
1b34ec43 DM	264	if (nla_put(skb, TCA_NAT_PARMS, sizeof(opt), &opt))
1b34ec43 DM	265	goto nla_put_failure;
48d8ee16 JHS	266
48d8ee16 JHS	267	tcf_tm_dump(&t, &p->tcf_tm);
9854518e	268	if (nla_put_64bit(skb, TCA_NAT_TM, sizeof(t), &t, TCA_NAT_PAD))
1b34ec43	269	goto nla_put_failure;
b4219952	270
b4219952 HX	271	return skb->len;
b4219952 HX	272
7ba699c6	273	nla_put_failure:
b4219952	274	nlmsg_trim(skb, b);
b4219952 HX	275	return -1;
	276	}
	277
ddf97ccd WC	278	static int tcf_nat_walker(struct net net, struct sk_buff skb,
ddf97ccd WC	279	struct netlink_callback *cb, int type,
a85a970a	280	const struct tc_action_ops *ops)
ddf97ccd WC	281	{
	282	struct tc_action_net *tn = net_generic(net, nat_net_id);
	283
a85a970a	284	return tcf_generic_walker(tn, skb, cb, type, ops);
ddf97ccd WC	285	}
ddf97ccd WC	286
a85a970a	287	static int tcf_nat_search(struct net net, struct tc_action *a, u32 index)
ddf97ccd WC	288	{
	289	struct tc_action_net *tn = net_generic(net, nat_net_id);
	290
65a206c0	291	return tcf_idr_search(tn, a, index);
ddf97ccd WC	292	}
ddf97ccd WC	293
b4219952 HX	294	static struct tc_action_ops act_nat_ops = {
b4219952 HX	295	.kind = "nat",
b4219952	296	.type = TCA_ACT_NAT,
b4219952 HX	297	.owner = THIS_MODULE,
	298	.act = tcf_nat,
	299	.dump = tcf_nat_dump,
b4219952	300	.init = tcf_nat_init,
ddf97ccd WC	301	.walk = tcf_nat_walker,
ddf97ccd WC	302	.lookup = tcf_nat_search,
a85a970a	303	.size = sizeof(struct tcf_nat),
ddf97ccd WC	304	};
	305
	306	static __net_init int nat_init_net(struct net *net)
	307	{
	308	struct tc_action_net *tn = net_generic(net, nat_net_id);
	309
65a206c0	310	return tc_action_net_init(tn, &act_nat_ops);
ddf97ccd WC	311	}
	312
	313	static void __net_exit nat_exit_net(struct net *net)
	314	{
	315	struct tc_action_net *tn = net_generic(net, nat_net_id);
	316
	317	tc_action_net_exit(tn);
	318	}
	319
	320	static struct pernet_operations nat_net_ops = {
	321	.init = nat_init_net,
	322	.exit = nat_exit_net,
	323	.id = &nat_net_id,
	324	.size = sizeof(struct tc_action_net),
b4219952 HX	325	};
	326
	327	MODULE_DESCRIPTION("Stateless NAT actions");
	328	MODULE_LICENSE("GPL");
	329
	330	static int __init nat_init_module(void)
	331	{
ddf97ccd	332	return tcf_register_action(&act_nat_ops, &nat_net_ops);
b4219952 HX	333	}
	334
	335	static void __exit nat_cleanup_module(void)
	336	{
ddf97ccd	337	tcf_unregister_action(&act_nat_ops, &nat_net_ops);
b4219952 HX	338	}
	339
	340	module_init(nat_init_module);
	341	module_exit(nat_cleanup_module);