[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)
{
	u_int16_t proto = conn->tuplehash[0].tuple.dst.protonum;

	if (proto == 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)
{
	struct nf_conntrack_tuple_hash *found;
	struct xt_connlimit_conn *conn;
	struct xt_connlimit_conn *tmp;
	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)];

	read_lock_bh(&nf_conntrack_lock);

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

	read_unlock_bh(&nf_conntrack_lock);

	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)
{
	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	{
	75	u_int16_t proto = conn->tuplehash[0].tuple.dst.protonum;
	76
	77	if (proto == IPPROTO_TCP)
	78	return conn->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT;
	79	else
	80	return 0;
	81	}
	82
	83	static inline unsigned int
643a2c15 JE	84	same_source_net(const union nf_inet_addr *addr,
	85	const union nf_inet_addr *mask,
	86	const union nf_inet_addr *u3, unsigned int family)
370786f9 JE	87	{
	88	if (family == AF_INET) {
	89	return (addr->ip & mask->ip) == (u3->ip & mask->ip);
	90	} else {
643a2c15	91	union nf_inet_addr lh, rh;
370786f9 JE	92	unsigned int i;
	93
	94	for (i = 0; i < ARRAY_SIZE(addr->ip6); ++i) {
	95	lh.ip6[i] = addr->ip6[i] & mask->ip6[i];
	96	rh.ip6[i] = u3->ip6[i] & mask->ip6[i];
	97	}
	98
	99	return memcmp(&lh.ip6, &rh.ip6, sizeof(lh.ip6)) == 0;
	100	}
	101	}
	102
	103	static int count_them(struct xt_connlimit_data *data,
	104	const struct nf_conntrack_tuple *tuple,
643a2c15 JE	105	const union nf_inet_addr *addr,
643a2c15 JE	106	const union nf_inet_addr *mask,
370786f9 JE	107	const struct xt_match *match)
	108	{
	109	struct nf_conntrack_tuple_hash *found;
	110	struct xt_connlimit_conn *conn;
	111	struct xt_connlimit_conn *tmp;
	112	struct nf_conn *found_ct;
	113	struct list_head *hash;
	114	bool addit = true;
	115	int matches = 0;
	116
	117
	118	if (match->family == AF_INET6)
	119	hash = &data->iphash[connlimit_iphash6(addr, mask)];
	120	else
	121	hash = &data->iphash[connlimit_iphash(addr->ip & mask->ip)];
	122
	123	read_lock_bh(&nf_conntrack_lock);
	124
	125	/* check the saved connections */
	126	list_for_each_entry_safe(conn, tmp, hash, list) {
	127	found = __nf_conntrack_find(&conn->tuple, NULL);
	128	found_ct = NULL;
	129
	130	if (found != NULL)
	131	found_ct = nf_ct_tuplehash_to_ctrack(found);
	132
	133	if (found_ct != NULL &&
	134	nf_ct_tuple_equal(&conn->tuple, tuple) &&
	135	!already_closed(found_ct))
	136	/*
	137	* Just to be sure we have it only once in the list.
	138	* We should not see tuples twice unless someone hooks
	139	* this into a table without "-p tcp --syn".
	140	*/
	141	addit = false;
	142
	143	if (found == NULL) {
	144	/* this one is gone */
	145	list_del(&conn->list);
	146	kfree(conn);
	147	continue;
	148	}
	149
	150	if (already_closed(found_ct)) {
	151	/*
	152	* we do not care about connections which are
	153	* closed already -> ditch it
	154	*/
	155	list_del(&conn->list);
	156	kfree(conn);
	157	continue;
	158	}
	159
	160	if (same_source_net(addr, mask, &conn->tuple.src.u3,
	161	match->family))
	162	/* same source network -> be counted! */
	163	++matches;
	164	}
	165
	166	read_unlock_bh(&nf_conntrack_lock);
	167
	168	if (addit) {
	169	/* save the new connection in our list */
	170	conn = kzalloc(sizeof(*conn), GFP_ATOMIC);
171	if (conn == NULL)
172	return -ENOMEM;
173	conn->tuple = *tuple;
174	list_add(&conn->list, hash);
175	++matches;
176	}
177
178	return matches;
179	}
180
d3c5ee6d JE	181	static bool
	182	connlimit_mt(const struct sk_buff skb, const struct net_device in,
	183	const struct net_device out, const struct xt_match match,
	184	const void *matchinfo, int offset, unsigned int protoff,
	185	bool *hotdrop)
370786f9 JE	186	{
370786f9 JE	187	const struct xt_connlimit_info *info = matchinfo;
22c2d8bc	188	union nf_inet_addr addr;
370786f9 JE	189	struct nf_conntrack_tuple tuple;
	190	const struct nf_conntrack_tuple *tuple_ptr = &tuple;
	191	enum ip_conntrack_info ctinfo;
	192	const struct nf_conn *ct;
	193	int connections;
	194
	195	ct = nf_ct_get(skb, &ctinfo);
	196	if (ct != NULL)
	197	tuple_ptr = &ct->tuplehash[0].tuple;
	198	else if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
	199	match->family, &tuple))
	200	goto hotdrop;
	201
	202	if (match->family == AF_INET6) {
	203	const struct ipv6hdr *iph = ipv6_hdr(skb);
	204	memcpy(&addr.ip6, &iph->saddr, sizeof(iph->saddr));
370786f9 JE	205	} else {
	206	const struct iphdr *iph = ip_hdr(skb);
	207	addr.ip = iph->saddr;
370786f9 JE	208	}
	209
	210	spin_lock_bh(&info->data->lock);
22c2d8bc JE	211	connections = count_them(info->data, tuple_ptr, &addr,
22c2d8bc JE	212	&info->mask, match);
370786f9 JE	213	spin_unlock_bh(&info->data->lock);
	214
	215	if (connections < 0) {
	216	/* kmalloc failed, drop it entirely */
	217	*hotdrop = true;
	218	return false;
	219	}
	220
	221	return (connections > info->limit) ^ info->inverse;
	222
	223	hotdrop:
	224	*hotdrop = true;
	225	return false;
	226	}
	227
d3c5ee6d JE	228	static bool
	229	connlimit_mt_check(const char tablename, const void ip,
	230	const struct xt_match match, void matchinfo,
	231	unsigned int hook_mask)
370786f9 JE	232	{
	233	struct xt_connlimit_info *info = matchinfo;
	234	unsigned int i;
	235
	236	if (nf_ct_l3proto_try_module_get(match->family) < 0) {
	237	printk(KERN_WARNING "cannot load conntrack support for "
	238	"address family %u\n", match->family);
	239	return false;
	240	}
	241
	242	/* init private data */
	243	info->data = kmalloc(sizeof(struct xt_connlimit_data), GFP_KERNEL);
	244	if (info->data == NULL) {
	245	nf_ct_l3proto_module_put(match->family);
	246	return false;
	247	}
	248
	249	spin_lock_init(&info->data->lock);
	250	for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i)
	251	INIT_LIST_HEAD(&info->data->iphash[i]);
	252
	253	return true;
	254	}
	255
d3c5ee6d JE	256	static void
d3c5ee6d JE	257	connlimit_mt_destroy(const struct xt_match match, void matchinfo)
370786f9 JE	258	{
	259	struct xt_connlimit_info *info = matchinfo;
	260	struct xt_connlimit_conn *conn;
	261	struct xt_connlimit_conn *tmp;
	262	struct list_head *hash = info->data->iphash;
	263	unsigned int i;
	264
	265	nf_ct_l3proto_module_put(match->family);
	266
	267	for (i = 0; i < ARRAY_SIZE(info->data->iphash); ++i) {
	268	list_for_each_entry_safe(conn, tmp, &hash[i], list) {
	269	list_del(&conn->list);
	270	kfree(conn);
	271	}
	272	}
	273
	274	kfree(info->data);
	275	}
	276
d3c5ee6d	277	static struct xt_match connlimit_mt_reg[] __read_mostly = {
370786f9 JE	278	{
	279	.name = "connlimit",
	280	.family = AF_INET,
d3c5ee6d JE	281	.checkentry = connlimit_mt_check,
d3c5ee6d JE	282	.match = connlimit_mt,
370786f9	283	.matchsize = sizeof(struct xt_connlimit_info),
d3c5ee6d	284	.destroy = connlimit_mt_destroy,
370786f9 JE	285	.me = THIS_MODULE,
	286	},
	287	{
	288	.name = "connlimit",
	289	.family = AF_INET6,
d3c5ee6d JE	290	.checkentry = connlimit_mt_check,
d3c5ee6d JE	291	.match = connlimit_mt,
370786f9	292	.matchsize = sizeof(struct xt_connlimit_info),
d3c5ee6d	293	.destroy = connlimit_mt_destroy,
370786f9 JE	294	.me = THIS_MODULE,
	295	},
	296	};
	297
d3c5ee6d	298	static int __init connlimit_mt_init(void)
370786f9	299	{
d3c5ee6d JE	300	return xt_register_matches(connlimit_mt_reg,
d3c5ee6d JE	301	ARRAY_SIZE(connlimit_mt_reg));
370786f9 JE	302	}
370786f9 JE	303
d3c5ee6d	304	static void __exit connlimit_mt_exit(void)
370786f9	305	{
d3c5ee6d	306	xt_unregister_matches(connlimit_mt_reg, ARRAY_SIZE(connlimit_mt_reg));
370786f9 JE	307	}
370786f9 JE	308
d3c5ee6d JE	309	module_init(connlimit_mt_init);
d3c5ee6d JE	310	module_exit(connlimit_mt_exit);
ba5dc275	311	MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>");
2ae15b64	312	MODULE_DESCRIPTION("Xtables: Number of connections matching");
370786f9 JE	313	MODULE_LICENSE("GPL");
	314	MODULE_ALIAS("ipt_connlimit");
	315	MODULE_ALIAS("ip6t_connlimit");