[deliverable/linux.git] / net / netfilter / xt_connlimit.c

/*
 * netfilter module to limit the number of parallel tcp
 * connections per IP address.
 *   (c) 2000 Gerd Knorr <kraxel@bytesex.org>
 *   Nov 2002: Martin Bene <martin.bene@icomedias.com>:
 *		only ignore TIME_WAIT or gone connections
 *   (C) CC Computer Consultants GmbH, 2007
 *   Contact: <jengelh@computergmbh.de>
 *
 * based on ...
 *
 * Kernel module to match connection tracking information.
 * GPL (C) 1999  Rusty Russell (rusty@rustcorp.com.au).
 */
#include <linux/in.h>
#include <linux/in6.h>
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/jhash.h>
#include <linux/list.h>
#include <linux/module.h>
#include <linux/random.h>
#include <linux/skbuff.h>
#include <linux/spinlock.h>
#include <linux/netfilter/nf_conntrack_tcp.h>
#include <linux/netfilter/x_tables.h>
#include <linux/netfilter/xt_connlimit.h>
#include <net/netfilter/nf_conntrack.h>
#include <net/netfilter/nf_conntrack_core.h>
#include <net/netfilter/nf_conntrack_tuple.h>

/* we will save the tuples of all connections we care about */
struct xt_connlimit_conn {
	struct list_head list;
	struct nf_conntrack_tuple tuple;
};

struct xt_connlimit_data {
	struct list_head iphash[256];
	spinlock_t lock;
};

static u_int32_t connlimit_rnd;
static bool connlimit_rnd_inited;

static inline unsigned int connlimit_iphash(__be32 addr)
{
	if (unlikely(!connlimit_rnd_inited)) {
		get_random_bytes(&connlimit_rnd, sizeof(connlimit_rnd));
		connlimit_rnd_inited = true;
	}
	return jhash_1word((__force __u32)addr, connlimit_rnd) & 0xFF;
}

static inline unsigned int
connlimit_iphash6(const union nf_inet_addr *addr,
                  const union nf_inet_addr *mask)
{
	union nf_inet_addr res;
	unsigned int i;

	if (unlikely(!connlimit_rnd_inited)) {
		get_random_bytes(&connlimit_rnd, sizeof(connlimit_rnd));
		connlimit_rnd_inited = true;
	}

	for (i = 0; i < ARRAY_SIZE(addr->ip6); ++i)
		res.ip6[i] = addr->ip6[i] & mask->ip6[i];

	return jhash2((u32 *)res.ip6, ARRAY_SIZE(res.ip6), connlimit_rnd) & 0xFF;
}

static inline bool already_closed(const struct nf_conn *conn)
{
	if (nf_ct_protonum(conn) == IPPROTO_TCP)
		return conn->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT;
	else
		return 0;
}

static inline unsigned int
same_source_net(const union nf_inet_addr *addr,
		const union nf_inet_addr *mask,
		const union nf_inet_addr *u3, unsigned int family)
{
	if (family == AF_INET) {
		return (addr->ip & mask->ip) == (u3->ip & mask->ip);
	} else {
		union nf_inet_addr lh, rh;
		unsigned int i;

		for (i = 0; i < ARRAY_SIZE(addr->ip6); ++i) {
			lh.ip6[i] = addr->ip6[i] & mask->ip6[i];
			rh.ip6[i] = u3->ip6[i] & mask->ip6[i];
		}

		return memcmp(&lh.ip6, &rh.ip6, sizeof(lh.ip6)) == 0;
	}
}

static int count_them(struct xt_connlimit_data *data,
		      const struct nf_conntrack_tuple *tuple,
		      const union nf_inet_addr *addr,
		      const union nf_inet_addr *mask,
		      const struct xt_match *match)
{
	const struct nf_conntrack_tuple_hash *found;
	struct xt_connlimit_conn *conn;
	struct xt_connlimit_conn *tmp;
	const struct nf_conn *found_ct;
	struct list_head *hash;
	bool addit = true;
	int matches = 0;


	if (match->family == AF_INET6)
		hash = &data->iphash[connlimit_iphash6(addr, mask)];
	else
		hash = &data->iphash[connlimit_iphash(addr->ip & mask->ip)];

	rcu_read_lock();

	/* check the saved connections */
	list_for_each_entry_safe(conn, tmp, hash, list) {
		found    = __nf_conntrack_find(&conn->tuple);
		found_ct = NULL;

		if (found != NULL)
			found_ct = nf_ct_tuplehash_to_ctrack(found);

		if (found_ct != NULL &&
		    nf_ct_tuple_equal(&conn->tuple, tuple) &&
		    !already_closed(found_ct))
			/*
			 * Just to be sure we have it only once in the list.
			 * We should not see tuples twice unless someone hooks
			 * this into a table without "-p tcp --syn".
			 */
			addit = false;

		if (found == NULL) {
			/* this one is gone */
			list_del(&conn->list);
			kfree(conn);
			continue;
		}

		if (already_closed(found_ct)) {
			/*
			 * we do not care about connections which are
			 * closed already -> ditch it
			 */
			list_del(&conn->list);
			kfree(conn);
			continue;
		}

		if (same_source_net(addr, mask, &conn->tuple.src.u3,
		    match->family))
			/* same source network -> be counted! */
			++matches;
	}

	rcu_read_unlock();

	if (addit) {
		/* save the new connection in our list */
		conn = kzalloc(sizeof(*conn), GFP_ATOMIC);
		if (conn == NULL)
			return -ENOMEM;
		conn->tuple = *tuple;
		list_add(&conn->list, hash);
		++matches;
	}

	return matches;
}

static bool
connlimit_mt(const struct sk_buff *skb, const struct net_device *in,
             const struct net_device *out, const struct xt_match *match,
             const void *matchinfo, int offset, unsigned int protoff,
             bool *hotdrop)
{
	const struct xt_connlimit_info *info = matchinfo;
	union nf_inet_addr addr;
	struct nf_conntrack_tuple tuple;
	const struct nf_conntrack_tuple *tuple_ptr = &tuple;
	enum ip_conntrack_info ctinfo;
	const struct nf_conn *ct;
	int connections;

	ct = nf_ct_get(skb, &ctinfo);
	if (ct != NULL)
		tuple_ptr = &ct->tuplehash[0].tuple;
	else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
				    match->family, &tuple))
		goto hotdrop;

	if (match->family == AF_INET6) {
		const struct ipv6hdr *iph = ipv6_hdr(skb);
		memcpy(&addr.ip6, &iph->saddr, sizeof(iph->saddr));
	} else {
		const struct iphdr *iph = ip_hdr(skb);
		addr.ip = iph->saddr;
	}

	spin_lock_bh(&info->data->lock);
	connections = count_them(info->data, tuple_ptr, &addr,
	                         &info->mask, match);
	spin_unlock_bh(&info->data->lock);

	if (connections < 0) {
		/* kmalloc failed, drop it entirely */
		*hotdrop = true;
		return false;
	}

	return (connections > info->limit) ^ info->inverse;

 hotdrop:
	*hotdrop = true;
	return false;
}

static bool
connlimit_mt_check(const char *tablename, const void *ip,
                   const struct xt_match *match, void *matchinfo,
                   unsigned int hook_mask)
{
	struct xt_connlimit_info *info = matchinfo;
	unsigned int i;

	if (nf_ct_l3proto_try_module_get(match->family) < 0) {
		printk(KERN_WARNING "cannot load conntrack support for "
		       "address family %u\n", match->family);
		return false;
	}

	/* init private data */
	info->data = kmalloc(sizeof(struct xt_connlimit_data), GFP_KERNEL);
	if (info->data == NULL) {
		nf_ct_l3proto_module_put(match->family);
		return false;
	}

	spin_lock_init(&info->data->lock);
	for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i)
		INIT_LIST_HEAD(&info->data->iphash[i]);

	return true;
}

static void
connlimit_mt_destroy(const struct xt_match *match, void *matchinfo)
{
	const struct xt_connlimit_info *info = matchinfo;
	struct xt_connlimit_conn *conn;
	struct xt_connlimit_conn *tmp;
	struct list_head *hash = info->data->iphash;
	unsigned int i;

	nf_ct_l3proto_module_put(match->family);

	for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i) {
		list_for_each_entry_safe(conn, tmp, &hash[i], list) {
			list_del(&conn->list);
			kfree(conn);
		}
	}

	kfree(info->data);
}

static struct xt_match connlimit_mt_reg[] __read_mostly = {
	{
		.name       = "connlimit",
		.family     = AF_INET,
		.checkentry = connlimit_mt_check,
		.match      = connlimit_mt,
		.matchsize  = sizeof(struct xt_connlimit_info),
		.destroy    = connlimit_mt_destroy,
		.me         = THIS_MODULE,
	},
	{
		.name       = "connlimit",
		.family     = AF_INET6,
		.checkentry = connlimit_mt_check,
		.match      = connlimit_mt,
		.matchsize  = sizeof(struct xt_connlimit_info),
		.destroy    = connlimit_mt_destroy,
		.me         = THIS_MODULE,
	},
};

static int __init connlimit_mt_init(void)
{
	return xt_register_matches(connlimit_mt_reg,
	       ARRAY_SIZE(connlimit_mt_reg));
}

static void __exit connlimit_mt_exit(void)
{
	xt_unregister_matches(connlimit_mt_reg, ARRAY_SIZE(connlimit_mt_reg));
}

module_init(connlimit_mt_init);
module_exit(connlimit_mt_exit);
MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>");
MODULE_DESCRIPTION("Xtables: Number of connections matching");
MODULE_LICENSE("GPL");
MODULE_ALIAS("ipt_connlimit");
MODULE_ALIAS("ip6t_connlimit");
Commit	Line	Data
370786f9 JE	1	/*
	2	* netfilter module to limit the number of parallel tcp
	3	* connections per IP address.
	4	* (c) 2000 Gerd Knorr <kraxel@bytesex.org>
	5	* Nov 2002: Martin Bene <martin.bene@icomedias.com>:
	6	* only ignore TIME_WAIT or gone connections
ba5dc275 JE	7	* (C) CC Computer Consultants GmbH, 2007
ba5dc275 JE	8	* Contact: <jengelh@computergmbh.de>
370786f9 JE	9	*
	10	* based on ...
	11	*
	12	* Kernel module to match connection tracking information.
	13	* GPL (C) 1999 Rusty Russell (rusty@rustcorp.com.au).
	14	*/
	15	#include <linux/in.h>
	16	#include <linux/in6.h>
	17	#include <linux/ip.h>
	18	#include <linux/ipv6.h>
	19	#include <linux/jhash.h>
	20	#include <linux/list.h>
	21	#include <linux/module.h>
	22	#include <linux/random.h>
	23	#include <linux/skbuff.h>
	24	#include <linux/spinlock.h>
	25	#include <linux/netfilter/nf_conntrack_tcp.h>
	26	#include <linux/netfilter/x_tables.h>
	27	#include <linux/netfilter/xt_connlimit.h>
	28	#include <net/netfilter/nf_conntrack.h>
	29	#include <net/netfilter/nf_conntrack_core.h>
	30	#include <net/netfilter/nf_conntrack_tuple.h>
	31
	32	/* we will save the tuples of all connections we care about */
	33	struct xt_connlimit_conn {
	34	struct list_head list;
	35	struct nf_conntrack_tuple tuple;
	36	};
	37
	38	struct xt_connlimit_data {
	39	struct list_head iphash[256];
	40	spinlock_t lock;
	41	};
	42
	43	static u_int32_t connlimit_rnd;
	44	static bool connlimit_rnd_inited;
	45
a34c4589	46	static inline unsigned int connlimit_iphash(__be32 addr)
370786f9 JE	47	{
	48	if (unlikely(!connlimit_rnd_inited)) {
	49	get_random_bytes(&connlimit_rnd, sizeof(connlimit_rnd));
	50	connlimit_rnd_inited = true;
	51	}
a34c4589	52	return jhash_1word((__force __u32)addr, connlimit_rnd) & 0xFF;
370786f9 JE	53	}
	54
	55	static inline unsigned int
643a2c15 JE	56	connlimit_iphash6(const union nf_inet_addr *addr,
643a2c15 JE	57	const union nf_inet_addr *mask)
370786f9	58	{
643a2c15	59	union nf_inet_addr res;
370786f9 JE	60	unsigned int i;
	61
	62	if (unlikely(!connlimit_rnd_inited)) {
	63	get_random_bytes(&connlimit_rnd, sizeof(connlimit_rnd));
	64	connlimit_rnd_inited = true;
	65	}
	66
	67	for (i = 0; i < ARRAY_SIZE(addr->ip6); ++i)
	68	res.ip6[i] = addr->ip6[i] & mask->ip6[i];
	69
a34c4589	70	return jhash2((u32 *)res.ip6, ARRAY_SIZE(res.ip6), connlimit_rnd) & 0xFF;
370786f9 JE	71	}
	72
	73	static inline bool already_closed(const struct nf_conn *conn)
	74	{
5e8fbe2a	75	if (nf_ct_protonum(conn) == IPPROTO_TCP)
370786f9 JE	76	return conn->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT;
	77	else
	78	return 0;
	79	}
	80
	81	static inline unsigned int
643a2c15 JE	82	same_source_net(const union nf_inet_addr *addr,
	83	const union nf_inet_addr *mask,
	84	const union nf_inet_addr *u3, unsigned int family)
370786f9 JE	85	{
	86	if (family == AF_INET) {
	87	return (addr->ip & mask->ip) == (u3->ip & mask->ip);
	88	} else {
643a2c15	89	union nf_inet_addr lh, rh;
370786f9 JE	90	unsigned int i;
	91
	92	for (i = 0; i < ARRAY_SIZE(addr->ip6); ++i) {
	93	lh.ip6[i] = addr->ip6[i] & mask->ip6[i];
	94	rh.ip6[i] = u3->ip6[i] & mask->ip6[i];
	95	}
	96
	97	return memcmp(&lh.ip6, &rh.ip6, sizeof(lh.ip6)) == 0;
	98	}
	99	}
	100
	101	static int count_them(struct xt_connlimit_data *data,
	102	const struct nf_conntrack_tuple *tuple,
643a2c15 JE	103	const union nf_inet_addr *addr,
643a2c15 JE	104	const union nf_inet_addr *mask,
370786f9 JE	105	const struct xt_match *match)
370786f9 JE	106	{
3cf93c96	107	const struct nf_conntrack_tuple_hash *found;
370786f9 JE	108	struct xt_connlimit_conn *conn;
370786f9 JE	109	struct xt_connlimit_conn *tmp;
3cf93c96	110	const struct nf_conn *found_ct;
370786f9 JE	111	struct list_head *hash;
	112	bool addit = true;
	113	int matches = 0;
	114
	115
	116	if (match->family == AF_INET6)
	117	hash = &data->iphash[connlimit_iphash6(addr, mask)];
	118	else
	119	hash = &data->iphash[connlimit_iphash(addr->ip & mask->ip)];
	120
76507f69	121	rcu_read_lock();
370786f9 JE	122
	123	/* check the saved connections */
	124	list_for_each_entry_safe(conn, tmp, hash, list) {
ba419aff	125	found = __nf_conntrack_find(&conn->tuple);
370786f9 JE	126	found_ct = NULL;
	127
	128	if (found != NULL)
	129	found_ct = nf_ct_tuplehash_to_ctrack(found);
	130
	131	if (found_ct != NULL &&
	132	nf_ct_tuple_equal(&conn->tuple, tuple) &&
	133	!already_closed(found_ct))
	134	/*
	135	* Just to be sure we have it only once in the list.
	136	* We should not see tuples twice unless someone hooks
	137	* this into a table without "-p tcp --syn".
	138	*/
	139	addit = false;
	140
	141	if (found == NULL) {
	142	/* this one is gone */
	143	list_del(&conn->list);
	144	kfree(conn);
	145	continue;
	146	}
	147
	148	if (already_closed(found_ct)) {
	149	/*
	150	* we do not care about connections which are
	151	* closed already -> ditch it
	152	*/
	153	list_del(&conn->list);
	154	kfree(conn);
	155	continue;
	156	}
	157
	158	if (same_source_net(addr, mask, &conn->tuple.src.u3,
	159	match->family))
	160	/* same source network -> be counted! */
	161	++matches;
	162	}
	163
76507f69	164	rcu_read_unlock();
370786f9 JE	165
	166	if (addit) {
	167	/* save the new connection in our list */
	168	conn = kzalloc(sizeof(*conn), GFP_ATOMIC);
	169	if (conn == NULL)
	170	return -ENOMEM;
	171	conn->tuple = *tuple;
	172	list_add(&conn->list, hash);
	173	++matches;
	174	}
	175
	176	return matches;
	177	}
	178
d3c5ee6d JE	179	static bool
	180	connlimit_mt(const struct sk_buff skb, const struct net_device in,
	181	const struct net_device out, const struct xt_match match,
	182	const void *matchinfo, int offset, unsigned int protoff,
	183	bool *hotdrop)
370786f9 JE	184	{
370786f9 JE	185	const struct xt_connlimit_info *info = matchinfo;
22c2d8bc	186	union nf_inet_addr addr;
370786f9 JE	187	struct nf_conntrack_tuple tuple;
	188	const struct nf_conntrack_tuple *tuple_ptr = &tuple;
	189	enum ip_conntrack_info ctinfo;
	190	const struct nf_conn *ct;
	191	int connections;
	192
	193	ct = nf_ct_get(skb, &ctinfo);
	194	if (ct != NULL)
	195	tuple_ptr = &ct->tuplehash[0].tuple;
	196	else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
	197	match->family, &tuple))
	198	goto hotdrop;
	199
	200	if (match->family == AF_INET6) {
	201	const struct ipv6hdr *iph = ipv6_hdr(skb);
	202	memcpy(&addr.ip6, &iph->saddr, sizeof(iph->saddr));
370786f9 JE	203	} else {
	204	const struct iphdr *iph = ip_hdr(skb);
	205	addr.ip = iph->saddr;
370786f9 JE	206	}
	207
	208	spin_lock_bh(&info->data->lock);
22c2d8bc JE	209	connections = count_them(info->data, tuple_ptr, &addr,
22c2d8bc JE	210	&info->mask, match);
370786f9 JE	211	spin_unlock_bh(&info->data->lock);
	212
	213	if (connections < 0) {
	214	/* kmalloc failed, drop it entirely */
	215	*hotdrop = true;
	216	return false;
	217	}
	218
	219	return (connections > info->limit) ^ info->inverse;
	220
	221	hotdrop:
	222	*hotdrop = true;
	223	return false;
	224	}
	225
d3c5ee6d JE	226	static bool
	227	connlimit_mt_check(const char tablename, const void ip,
	228	const struct xt_match match, void matchinfo,
	229	unsigned int hook_mask)
370786f9 JE	230	{
	231	struct xt_connlimit_info *info = matchinfo;
	232	unsigned int i;
	233
	234	if (nf_ct_l3proto_try_module_get(match->family) < 0) {
	235	printk(KERN_WARNING "cannot load conntrack support for "
	236	"address family %u\n", match->family);
	237	return false;
	238	}
	239
	240	/* init private data */
	241	info->data = kmalloc(sizeof(struct xt_connlimit_data), GFP_KERNEL);
	242	if (info->data == NULL) {
	243	nf_ct_l3proto_module_put(match->family);
	244	return false;
	245	}
	246
	247	spin_lock_init(&info->data->lock);
	248	for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i)
	249	INIT_LIST_HEAD(&info->data->iphash[i]);
	250
	251	return true;
	252	}
	253
d3c5ee6d JE	254	static void
d3c5ee6d JE	255	connlimit_mt_destroy(const struct xt_match match, void matchinfo)
370786f9	256	{
3cf93c96	257	const struct xt_connlimit_info *info = matchinfo;
370786f9 JE	258	struct xt_connlimit_conn *conn;
	259	struct xt_connlimit_conn *tmp;
	260	struct list_head *hash = info->data->iphash;
	261	unsigned int i;
	262
	263	nf_ct_l3proto_module_put(match->family);
	264
	265	for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i) {
	266	list_for_each_entry_safe(conn, tmp, &hash[i], list) {
	267	list_del(&conn->list);
	268	kfree(conn);
	269	}
	270	}
	271
	272	kfree(info->data);
	273	}
	274
d3c5ee6d	275	static struct xt_match connlimit_mt_reg[] __read_mostly = {
370786f9 JE	276	{
	277	.name = "connlimit",
	278	.family = AF_INET,
d3c5ee6d JE	279	.checkentry = connlimit_mt_check,
d3c5ee6d JE	280	.match = connlimit_mt,
370786f9	281	.matchsize = sizeof(struct xt_connlimit_info),
d3c5ee6d	282	.destroy = connlimit_mt_destroy,
370786f9 JE	283	.me = THIS_MODULE,
	284	},
	285	{
	286	.name = "connlimit",
	287	.family = AF_INET6,
d3c5ee6d JE	288	.checkentry = connlimit_mt_check,
d3c5ee6d JE	289	.match = connlimit_mt,
370786f9	290	.matchsize = sizeof(struct xt_connlimit_info),
d3c5ee6d	291	.destroy = connlimit_mt_destroy,
370786f9 JE	292	.me = THIS_MODULE,
	293	},
	294	};
	295
d3c5ee6d	296	static int __init connlimit_mt_init(void)
370786f9	297	{
d3c5ee6d JE	298	return xt_register_matches(connlimit_mt_reg,
d3c5ee6d JE	299	ARRAY_SIZE(connlimit_mt_reg));
370786f9 JE	300	}
370786f9 JE	301
d3c5ee6d	302	static void __exit connlimit_mt_exit(void)
370786f9	303	{
d3c5ee6d	304	xt_unregister_matches(connlimit_mt_reg, ARRAY_SIZE(connlimit_mt_reg));
370786f9 JE	305	}
370786f9 JE	306
d3c5ee6d JE	307	module_init(connlimit_mt_init);
d3c5ee6d JE	308	module_exit(connlimit_mt_exit);
ba5dc275	309	MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>");
2ae15b64	310	MODULE_DESCRIPTION("Xtables: Number of connections matching");
370786f9 JE	311	MODULE_LICENSE("GPL");
	312	MODULE_ALIAS("ipt_connlimit");
	313	MODULE_ALIAS("ip6t_connlimit");