[linux-block.git] / net / netfilter / xt_socket.c

/*
 * Transparent proxy support for Linux/iptables
 *
 * Copyright (C) 2007-2008 BalaBit IT Ltd.
 * Author: Krisztian Kovacs
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 *
 */
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/netfilter/x_tables.h>
#include <linux/netfilter_ipv4/ip_tables.h>
#include <net/tcp.h>
#include <net/udp.h>
#include <net/icmp.h>
#include <net/sock.h>
#include <net/inet_sock.h>
#include <net/netfilter/ipv4/nf_defrag_ipv4.h>

#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
#define XT_SOCKET_HAVE_IPV6 1
#include <linux/netfilter_ipv6/ip6_tables.h>
#include <net/inet6_hashtables.h>
#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
#endif

#include <linux/netfilter/xt_socket.h>

#if IS_ENABLED(CONFIG_NF_CONNTRACK)
#define XT_SOCKET_HAVE_CONNTRACK 1
#include <net/netfilter/nf_conntrack.h>
#endif

static int
extract_icmp4_fields(const struct sk_buff *skb,
		    u8 *protocol,
		    __be32 *raddr,
		    __be32 *laddr,
		    __be16 *rport,
		    __be16 *lport)
{
	unsigned int outside_hdrlen = ip_hdrlen(skb);
	struct iphdr *inside_iph, _inside_iph;
	struct icmphdr *icmph, _icmph;
	__be16 *ports, _ports[2];

	icmph = skb_header_pointer(skb, outside_hdrlen,
				   sizeof(_icmph), &_icmph);
	if (icmph == NULL)
		return 1;

	switch (icmph->type) {
	case ICMP_DEST_UNREACH:
	case ICMP_SOURCE_QUENCH:
	case ICMP_REDIRECT:
	case ICMP_TIME_EXCEEDED:
	case ICMP_PARAMETERPROB:
		break;
	default:
		return 1;
	}

	inside_iph = skb_header_pointer(skb, outside_hdrlen +
					sizeof(struct icmphdr),
					sizeof(_inside_iph), &_inside_iph);
	if (inside_iph == NULL)
		return 1;

	if (inside_iph->protocol != IPPROTO_TCP &&
	    inside_iph->protocol != IPPROTO_UDP)
		return 1;

	ports = skb_header_pointer(skb, outside_hdrlen +
				   sizeof(struct icmphdr) +
				   (inside_iph->ihl << 2),
				   sizeof(_ports), &_ports);
	if (ports == NULL)
		return 1;

	/* the inside IP packet is the one quoted from our side, thus
	 * its saddr is the local address */
	*protocol = inside_iph->protocol;
	*laddr = inside_iph->saddr;
	*lport = ports[0];
	*raddr = inside_iph->daddr;
	*rport = ports[1];

	return 0;
}

/* "socket" match based redirection (no specific rule)
 * ===================================================
 *
 * There are connections with dynamic endpoints (e.g. FTP data
 * connection) that the user is unable to add explicit rules
 * for. These are taken care of by a generic "socket" rule. It is
 * assumed that the proxy application is trusted to open such
 * connections without explicit iptables rule (except of course the
 * generic 'socket' rule). In this case the following sockets are
 * matched in preference order:
 *
 *   - match: if there's a fully established connection matching the
 *     _packet_ tuple
 *
 *   - match: if there's a non-zero bound listener (possibly with a
 *     non-local address) We don't accept zero-bound listeners, since
 *     then local services could intercept traffic going through the
 *     box.
 */
static struct sock *
xt_socket_get_sock_v4(struct net *net, const u8 protocol,
		      const __be32 saddr, const __be32 daddr,
		      const __be16 sport, const __be16 dport,
		      const struct net_device *in)
{
	switch (protocol) {
	case IPPROTO_TCP:
		return __inet_lookup(net, &tcp_hashinfo,
				     saddr, sport, daddr, dport,
				     in->ifindex);
	case IPPROTO_UDP:
		return udp4_lib_lookup(net, saddr, sport, daddr, dport,
				       in->ifindex);
	}
	return NULL;
}

static bool xt_socket_sk_is_transparent(struct sock *sk)
{
	switch (sk->sk_state) {
	case TCP_TIME_WAIT:
		return inet_twsk(sk)->tw_transparent;

	case TCP_NEW_SYN_RECV:
		return inet_rsk(inet_reqsk(sk))->no_srccheck;

	default:
		return inet_sk(sk)->transparent;
	}
}

static struct sock *xt_socket_lookup_slow_v4(const struct sk_buff *skb,
					     const struct net_device *indev)
{
	const struct iphdr *iph = ip_hdr(skb);
	__be32 uninitialized_var(daddr), uninitialized_var(saddr);
	__be16 uninitialized_var(dport), uninitialized_var(sport);
	u8 uninitialized_var(protocol);
#ifdef XT_SOCKET_HAVE_CONNTRACK
	struct nf_conn const *ct;
	enum ip_conntrack_info ctinfo;
#endif

	if (iph->protocol == IPPROTO_UDP || iph->protocol == IPPROTO_TCP) {
		struct udphdr _hdr, *hp;

		hp = skb_header_pointer(skb, ip_hdrlen(skb),
					sizeof(_hdr), &_hdr);
		if (hp == NULL)
			return NULL;

		protocol = iph->protocol;
		saddr = iph->saddr;
		sport = hp->source;
		daddr = iph->daddr;
		dport = hp->dest;

	} else if (iph->protocol == IPPROTO_ICMP) {
		if (extract_icmp4_fields(skb, &protocol, &saddr, &daddr,
					 &sport, &dport))
			return NULL;
	} else {
		return NULL;
	}

#ifdef XT_SOCKET_HAVE_CONNTRACK
	/* Do the lookup with the original socket address in
	 * case this is a reply packet of an established
	 * SNAT-ted connection.
	 */
	ct = nf_ct_get(skb, &ctinfo);
	if (ct && !nf_ct_is_untracked(ct) &&
	    ((iph->protocol != IPPROTO_ICMP &&
	      ctinfo == IP_CT_ESTABLISHED_REPLY) ||
	     (iph->protocol == IPPROTO_ICMP &&
	      ctinfo == IP_CT_RELATED_REPLY)) &&
	    (ct->status & IPS_SRC_NAT_DONE)) {

		daddr = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip;
		dport = (iph->protocol == IPPROTO_TCP) ?
			ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.tcp.port :
			ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.udp.port;
	}
#endif

	return xt_socket_get_sock_v4(dev_net(skb->dev), protocol, saddr, daddr,
				     sport, dport, indev);
}

static bool
socket_match(const struct sk_buff *skb, struct xt_action_param *par,
	     const struct xt_socket_mtinfo1 *info)
{
	struct sock *sk = skb->sk;

	if (!sk)
		sk = xt_socket_lookup_slow_v4(skb, par->in);
	if (sk) {
		bool wildcard;
		bool transparent = true;

		/* Ignore sockets listening on INADDR_ANY,
		 * unless XT_SOCKET_NOWILDCARD is set
		 */
		wildcard = (!(info->flags & XT_SOCKET_NOWILDCARD) &&
			    sk_fullsock(sk) &&
			    inet_sk(sk)->inet_rcv_saddr == 0);

		/* Ignore non-transparent sockets,
		 * if XT_SOCKET_TRANSPARENT is used
		 */
		if (info->flags & XT_SOCKET_TRANSPARENT)
			transparent = xt_socket_sk_is_transparent(sk);

		if (sk != skb->sk)
			sock_gen_put(sk);

		if (wildcard || !transparent)
			sk = NULL;
	}

	return sk != NULL;
}

static bool
socket_mt4_v0(const struct sk_buff *skb, struct xt_action_param *par)
{
	static struct xt_socket_mtinfo1 xt_info_v0 = {
		.flags = 0,
	};

	return socket_match(skb, par, &xt_info_v0);
}

static bool
socket_mt4_v1_v2(const struct sk_buff *skb, struct xt_action_param *par)
{
	return socket_match(skb, par, par->matchinfo);
}

#ifdef XT_SOCKET_HAVE_IPV6

static int
extract_icmp6_fields(const struct sk_buff *skb,
		     unsigned int outside_hdrlen,
		     int *protocol,
		     const struct in6_addr **raddr,
		     const struct in6_addr **laddr,
		     __be16 *rport,
		     __be16 *lport,
		     struct ipv6hdr *ipv6_var)
{
	const struct ipv6hdr *inside_iph;
	struct icmp6hdr *icmph, _icmph;
	__be16 *ports, _ports[2];
	u8 inside_nexthdr;
	__be16 inside_fragoff;
	int inside_hdrlen;

	icmph = skb_header_pointer(skb, outside_hdrlen,
				   sizeof(_icmph), &_icmph);
	if (icmph == NULL)
		return 1;

	if (icmph->icmp6_type & ICMPV6_INFOMSG_MASK)
		return 1;

	inside_iph = skb_header_pointer(skb, outside_hdrlen + sizeof(_icmph),
					sizeof(*ipv6_var), ipv6_var);
	if (inside_iph == NULL)
		return 1;
	inside_nexthdr = inside_iph->nexthdr;

	inside_hdrlen = ipv6_skip_exthdr(skb, outside_hdrlen + sizeof(_icmph) +
					      sizeof(*ipv6_var),
					 &inside_nexthdr, &inside_fragoff);
	if (inside_hdrlen < 0)
		return 1; /* hjm: Packet has no/incomplete transport layer headers. */

	if (inside_nexthdr != IPPROTO_TCP &&
	    inside_nexthdr != IPPROTO_UDP)
		return 1;

	ports = skb_header_pointer(skb, inside_hdrlen,
				   sizeof(_ports), &_ports);
	if (ports == NULL)
		return 1;

	/* the inside IP packet is the one quoted from our side, thus
	 * its saddr is the local address */
	*protocol = inside_nexthdr;
	*laddr = &inside_iph->saddr;
	*lport = ports[0];
	*raddr = &inside_iph->daddr;
	*rport = ports[1];

	return 0;
}

static struct sock *
xt_socket_get_sock_v6(struct net *net, const u8 protocol,
		      const struct in6_addr *saddr, const struct in6_addr *daddr,
		      const __be16 sport, const __be16 dport,
		      const struct net_device *in)
{
	switch (protocol) {
	case IPPROTO_TCP:
		return inet6_lookup(net, &tcp_hashinfo,
				    saddr, sport, daddr, dport,
				    in->ifindex);
	case IPPROTO_UDP:
		return udp6_lib_lookup(net, saddr, sport, daddr, dport,
				       in->ifindex);
	}

	return NULL;
}

static struct sock *xt_socket_lookup_slow_v6(const struct sk_buff *skb,
					     const struct net_device *indev)
{
	__be16 uninitialized_var(dport), uninitialized_var(sport);
	const struct in6_addr *daddr = NULL, *saddr = NULL;
	struct ipv6hdr *iph = ipv6_hdr(skb);
	int thoff = 0, tproto;

	tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL);
	if (tproto < 0) {
		pr_debug("unable to find transport header in IPv6 packet, dropping\n");
		return NULL;
	}

	if (tproto == IPPROTO_UDP || tproto == IPPROTO_TCP) {
		struct udphdr _hdr, *hp;

		hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr);
		if (hp == NULL)
			return NULL;

		saddr = &iph->saddr;
		sport = hp->source;
		daddr = &iph->daddr;
		dport = hp->dest;

	} else if (tproto == IPPROTO_ICMPV6) {
		struct ipv6hdr ipv6_var;

		if (extract_icmp6_fields(skb, thoff, &tproto, &saddr, &daddr,
					 &sport, &dport, &ipv6_var))
			return NULL;
	} else {
		return NULL;
	}

	return xt_socket_get_sock_v6(dev_net(skb->dev), tproto, saddr, daddr,
				     sport, dport, indev);
}

static bool
socket_mt6_v1_v2(const struct sk_buff *skb, struct xt_action_param *par)
{
	const struct xt_socket_mtinfo1 *info = (struct xt_socket_mtinfo1 *) par->matchinfo;
	struct sock *sk = skb->sk;

	if (!sk)
		sk = xt_socket_lookup_slow_v6(skb, par->in);
	if (sk) {
		bool wildcard;
		bool transparent = true;

		/* Ignore sockets listening on INADDR_ANY
		 * unless XT_SOCKET_NOWILDCARD is set
		 */
		wildcard = (!(info->flags & XT_SOCKET_NOWILDCARD) &&
			    sk_fullsock(sk) &&
			    ipv6_addr_any(&sk->sk_v6_rcv_saddr));

		/* Ignore non-transparent sockets,
		 * if XT_SOCKET_TRANSPARENT is used
		 */
		if (info->flags & XT_SOCKET_TRANSPARENT)
			transparent = xt_socket_sk_is_transparent(sk);

		if (sk != skb->sk)
			sock_gen_put(sk);

		if (wildcard || !transparent)
			sk = NULL;
	}

	return sk != NULL;
}
#endif

static int socket_mt_v1_check(const struct xt_mtchk_param *par)
{
	const struct xt_socket_mtinfo1 *info = (struct xt_socket_mtinfo1 *) par->matchinfo;

	if (info->flags & ~XT_SOCKET_FLAGS_V1) {
		pr_info("unknown flags 0x%x\n", info->flags & ~XT_SOCKET_FLAGS_V1);
		return -EINVAL;
	}
	return 0;
}

static int socket_mt_v2_check(const struct xt_mtchk_param *par)
{
	const struct xt_socket_mtinfo2 *info = (struct xt_socket_mtinfo2 *) par->matchinfo;

	if (info->flags & ~XT_SOCKET_FLAGS_V2) {
		pr_info("unknown flags 0x%x\n", info->flags & ~XT_SOCKET_FLAGS_V2);
		return -EINVAL;
	}
	return 0;
}

static struct xt_match socket_mt_reg[] __read_mostly = {
	{
		.name		= "socket",
		.revision	= 0,
		.family		= NFPROTO_IPV4,
		.match		= socket_mt4_v0,
		.hooks		= (1 << NF_INET_PRE_ROUTING) |
				  (1 << NF_INET_LOCAL_IN),
		.me		= THIS_MODULE,
	},
	{
		.name		= "socket",
		.revision	= 1,
		.family		= NFPROTO_IPV4,
		.match		= socket_mt4_v1_v2,
		.checkentry	= socket_mt_v1_check,
		.matchsize	= sizeof(struct xt_socket_mtinfo1),
		.hooks		= (1 << NF_INET_PRE_ROUTING) |
				  (1 << NF_INET_LOCAL_IN),
		.me		= THIS_MODULE,
	},
#ifdef XT_SOCKET_HAVE_IPV6
	{
		.name		= "socket",
		.revision	= 1,
		.family		= NFPROTO_IPV6,
		.match		= socket_mt6_v1_v2,
		.checkentry	= socket_mt_v1_check,
		.matchsize	= sizeof(struct xt_socket_mtinfo1),
		.hooks		= (1 << NF_INET_PRE_ROUTING) |
				  (1 << NF_INET_LOCAL_IN),
		.me		= THIS_MODULE,
	},
#endif
	{
		.name		= "socket",
		.revision	= 2,
		.family		= NFPROTO_IPV4,
		.match		= socket_mt4_v1_v2,
		.checkentry	= socket_mt_v2_check,
		.matchsize	= sizeof(struct xt_socket_mtinfo1),
		.hooks		= (1 << NF_INET_PRE_ROUTING) |
				  (1 << NF_INET_LOCAL_IN),
		.me		= THIS_MODULE,
	},
#ifdef XT_SOCKET_HAVE_IPV6
	{
		.name		= "socket",
		.revision	= 2,
		.family		= NFPROTO_IPV6,
		.match		= socket_mt6_v1_v2,
		.checkentry	= socket_mt_v2_check,
		.matchsize	= sizeof(struct xt_socket_mtinfo1),
		.hooks		= (1 << NF_INET_PRE_ROUTING) |
				  (1 << NF_INET_LOCAL_IN),
		.me		= THIS_MODULE,
	},
#endif
};

static int __init socket_mt_init(void)
{
	nf_defrag_ipv4_enable();
#ifdef XT_SOCKET_HAVE_IPV6
	nf_defrag_ipv6_enable();
#endif

	return xt_register_matches(socket_mt_reg, ARRAY_SIZE(socket_mt_reg));
}

static void __exit socket_mt_exit(void)
{
	xt_unregister_matches(socket_mt_reg, ARRAY_SIZE(socket_mt_reg));
}

module_init(socket_mt_init);
module_exit(socket_mt_exit);

MODULE_LICENSE("GPL");
MODULE_AUTHOR("Krisztian Kovacs, Balazs Scheidler");
MODULE_DESCRIPTION("x_tables socket match module");
MODULE_ALIAS("ipt_socket");
MODULE_ALIAS("ip6t_socket");
Commit	Line	Data
136cdc71 KK	1	/*
	2	* Transparent proxy support for Linux/iptables
	3	*
	4	* Copyright (C) 2007-2008 BalaBit IT Ltd.
	5	* Author: Krisztian Kovacs
	6	*
	7	* This program is free software; you can redistribute it and/or modify
	8	* it under the terms of the GNU General Public License version 2 as
	9	* published by the Free Software Foundation.
	10	*
	11	*/
ff67e4e4	12	#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
136cdc71 KK	13	#include <linux/module.h>
	14	#include <linux/skbuff.h>
	15	#include <linux/netfilter/x_tables.h>
	16	#include <linux/netfilter_ipv4/ip_tables.h>
	17	#include <net/tcp.h>
	18	#include <net/udp.h>
	19	#include <net/icmp.h>
	20	#include <net/sock.h>
	21	#include <net/inet_sock.h>
136cdc71	22	#include <net/netfilter/ipv4/nf_defrag_ipv4.h>
f6318e55	23
c0cd1156	24	#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
f6318e55 KK	25	#define XT_SOCKET_HAVE_IPV6 1
f6318e55 KK	26	#include <linux/netfilter_ipv6/ip6_tables.h>
93742cf8	27	#include <net/inet6_hashtables.h>
b64c9256	28	#include <net/netfilter/ipv6/nf_defrag_ipv6.h>
f6318e55	29	#endif
136cdc71	30
a31e1ffd LAT	31	#include <linux/netfilter/xt_socket.h>
a31e1ffd LAT	32
c0cd1156	33	#if IS_ENABLED(CONFIG_NF_CONNTRACK)
136cdc71 KK	34	#define XT_SOCKET_HAVE_CONNTRACK 1
	35	#include <net/netfilter/nf_conntrack.h>
	36	#endif
	37
	38	static int
b64c9256	39	extract_icmp4_fields(const struct sk_buff *skb,
136cdc71 KK	40	u8 *protocol,
	41	__be32 *raddr,
	42	__be32 *laddr,
	43	__be16 *rport,
	44	__be16 *lport)
	45	{
	46	unsigned int outside_hdrlen = ip_hdrlen(skb);
	47	struct iphdr *inside_iph, _inside_iph;
	48	struct icmphdr *icmph, _icmph;
	49	__be16 *ports, _ports[2];
	50
	51	icmph = skb_header_pointer(skb, outside_hdrlen,
	52	sizeof(_icmph), &_icmph);
	53	if (icmph == NULL)
	54	return 1;
	55
	56	switch (icmph->type) {
	57	case ICMP_DEST_UNREACH:
	58	case ICMP_SOURCE_QUENCH:
	59	case ICMP_REDIRECT:
	60	case ICMP_TIME_EXCEEDED:
	61	case ICMP_PARAMETERPROB:
	62	break;
	63	default:
	64	return 1;
	65	}
	66
	67	inside_iph = skb_header_pointer(skb, outside_hdrlen +
	68	sizeof(struct icmphdr),
	69	sizeof(_inside_iph), &_inside_iph);
	70	if (inside_iph == NULL)
	71	return 1;
	72
	73	if (inside_iph->protocol != IPPROTO_TCP &&
	74	inside_iph->protocol != IPPROTO_UDP)
	75	return 1;
	76
	77	ports = skb_header_pointer(skb, outside_hdrlen +
	78	sizeof(struct icmphdr) +
	79	(inside_iph->ihl << 2),
	80	sizeof(_ports), &_ports);
	81	if (ports == NULL)
	82	return 1;
	83
	84	/* the inside IP packet is the one quoted from our side, thus
	85	* its saddr is the local address */
	86	*protocol = inside_iph->protocol;
	87	*laddr = inside_iph->saddr;
	88	*lport = ports[0];
	89	*raddr = inside_iph->daddr;
	90	*rport = ports[1];
	91
	92	return 0;
	93	}
	94
93742cf8 FW	95	/* "socket" match based redirection (no specific rule)
	96	* ===================================================
	97	*
	98	* There are connections with dynamic endpoints (e.g. FTP data
	99	* connection) that the user is unable to add explicit rules
	100	* for. These are taken care of by a generic "socket" rule. It is
	101	* assumed that the proxy application is trusted to open such
	102	* connections without explicit iptables rule (except of course the
	103	* generic 'socket' rule). In this case the following sockets are
	104	* matched in preference order:
	105	*
	106	* - match: if there's a fully established connection matching the
	107	* _packet_ tuple
	108	*
	109	* - match: if there's a non-zero bound listener (possibly with a
	110	* non-local address) We don't accept zero-bound listeners, since
	111	* then local services could intercept traffic going through the
	112	* box.
	113	*/
	114	static struct sock *
	115	xt_socket_get_sock_v4(struct net *net, const u8 protocol,
	116	const __be32 saddr, const __be32 daddr,
	117	const __be16 sport, const __be16 dport,
	118	const struct net_device *in)
	119	{
	120	switch (protocol) {
	121	case IPPROTO_TCP:
	122	return __inet_lookup(net, &tcp_hashinfo,
	123	saddr, sport, daddr, dport,
	124	in->ifindex);
	125	case IPPROTO_UDP:
	126	return udp4_lib_lookup(net, saddr, sport, daddr, dport,
	127	in->ifindex);
	128	}
	129	return NULL;
	130	}
	131
a9407000 ED	132	static bool xt_socket_sk_is_transparent(struct sock *sk)
	133	{
	134	switch (sk->sk_state) {
	135	case TCP_TIME_WAIT:
	136	return inet_twsk(sk)->tw_transparent;
	137
	138	case TCP_NEW_SYN_RECV:
	139	return inet_rsk(inet_reqsk(sk))->no_srccheck;
	140
	141	default:
	142	return inet_sk(sk)->transparent;
	143	}
	144	}
	145
d64d80a2 DB	146	static struct sock xt_socket_lookup_slow_v4(const struct sk_buff skb,
d64d80a2 DB	147	const struct net_device *indev)
136cdc71 KK	148	{
136cdc71 KK	149	const struct iphdr *iph = ip_hdr(skb);
6703aa74 PNA	150	__be32 uninitialized_var(daddr), uninitialized_var(saddr);
	151	__be16 uninitialized_var(dport), uninitialized_var(sport);
	152	u8 uninitialized_var(protocol);
136cdc71 KK	153	#ifdef XT_SOCKET_HAVE_CONNTRACK
	154	struct nf_conn const *ct;
	155	enum ip_conntrack_info ctinfo;
	156	#endif
	157
	158	if (iph->protocol == IPPROTO_UDP \|\| iph->protocol == IPPROTO_TCP) {
d64d80a2 DB	159	struct udphdr _hdr, *hp;
d64d80a2 DB	160
136cdc71 KK	161	hp = skb_header_pointer(skb, ip_hdrlen(skb),
	162	sizeof(_hdr), &_hdr);
	163	if (hp == NULL)
d64d80a2	164	return NULL;
136cdc71 KK	165
	166	protocol = iph->protocol;
	167	saddr = iph->saddr;
	168	sport = hp->source;
	169	daddr = iph->daddr;
	170	dport = hp->dest;
	171
	172	} else if (iph->protocol == IPPROTO_ICMP) {
b64c9256	173	if (extract_icmp4_fields(skb, &protocol, &saddr, &daddr,
d64d80a2 DB	174	&sport, &dport))
d64d80a2 DB	175	return NULL;
136cdc71	176	} else {
d64d80a2	177	return NULL;
136cdc71 KK	178	}
	179
	180	#ifdef XT_SOCKET_HAVE_CONNTRACK
d64d80a2 DB	181	/* Do the lookup with the original socket address in
	182	* case this is a reply packet of an established
	183	* SNAT-ted connection.
	184	*/
136cdc71	185	ct = nf_ct_get(skb, &ctinfo);
5bfddbd4	186	if (ct && !nf_ct_is_untracked(ct) &&
136cdc71	187	((iph->protocol != IPPROTO_ICMP &&
fb048833	188	ctinfo == IP_CT_ESTABLISHED_REPLY) \|\|
136cdc71	189	(iph->protocol == IPPROTO_ICMP &&
fb048833	190	ctinfo == IP_CT_RELATED_REPLY)) &&
136cdc71 KK	191	(ct->status & IPS_SRC_NAT_DONE)) {
	192
	193	daddr = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip;
	194	dport = (iph->protocol == IPPROTO_TCP) ?
	195	ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.tcp.port :
	196	ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.udp.port;
	197	}
	198	#endif
	199
d64d80a2 DB	200	return xt_socket_get_sock_v4(dev_net(skb->dev), protocol, saddr, daddr,
	201	sport, dport, indev);
	202	}
	203
	204	static bool
	205	socket_match(const struct sk_buff skb, struct xt_action_param par,
	206	const struct xt_socket_mtinfo1 *info)
	207	{
	208	struct sock *sk = skb->sk;
	209
00028aa3	210	if (!sk)
d64d80a2	211	sk = xt_socket_lookup_slow_v4(skb, par->in);
00028aa3	212	if (sk) {
a31e1ffd LAT	213	bool wildcard;
	214	bool transparent = true;
	215
681f130f ED	216	/* Ignore sockets listening on INADDR_ANY,
	217	* unless XT_SOCKET_NOWILDCARD is set
	218	*/
	219	wildcard = (!(info->flags & XT_SOCKET_NOWILDCARD) &&
a9407000	220	sk_fullsock(sk) &&
c720c7e8	221	inet_sk(sk)->inet_rcv_saddr == 0);
a31e1ffd LAT	222
a31e1ffd LAT	223	/* Ignore non-transparent sockets,
a9407000 ED	224	* if XT_SOCKET_TRANSPARENT is used
a9407000 ED	225	*/
baf60efa	226	if (info->flags & XT_SOCKET_TRANSPARENT)
a9407000	227	transparent = xt_socket_sk_is_transparent(sk);
136cdc71	228
00028aa3	229	if (sk != skb->sk)
1a8bf6ee	230	sock_gen_put(sk);
a31e1ffd LAT	231
a31e1ffd LAT	232	if (wildcard \|\| !transparent)
136cdc71 KK	233	sk = NULL;
	234	}
	235
d64d80a2	236	return sk != NULL;
136cdc71 KK	237	}
136cdc71 KK	238
a31e1ffd	239	static bool
b64c9256	240	socket_mt4_v0(const struct sk_buff skb, struct xt_action_param par)
a31e1ffd	241	{
baf60efa ED	242	static struct xt_socket_mtinfo1 xt_info_v0 = {
	243	.flags = 0,
	244	};
	245
	246	return socket_match(skb, par, &xt_info_v0);
a31e1ffd LAT	247	}
	248
	249	static bool
681f130f	250	socket_mt4_v1_v2(const struct sk_buff skb, struct xt_action_param par)
a31e1ffd LAT	251	{
	252	return socket_match(skb, par, par->matchinfo);
	253	}
	254
f6318e55	255	#ifdef XT_SOCKET_HAVE_IPV6
b64c9256 BS	256
	257	static int
	258	extract_icmp6_fields(const struct sk_buff *skb,
	259	unsigned int outside_hdrlen,
089282fb	260	int *protocol,
78296c97 ED	261	const struct in6_addr **raddr,
78296c97 ED	262	const struct in6_addr **laddr,
b64c9256	263	__be16 *rport,
78296c97 ED	264	__be16 *lport,
78296c97 ED	265	struct ipv6hdr *ipv6_var)
b64c9256	266	{
78296c97	267	const struct ipv6hdr *inside_iph;
b64c9256 BS	268	struct icmp6hdr *icmph, _icmph;
	269	__be16 *ports, _ports[2];
	270	u8 inside_nexthdr;
75f2811c	271	__be16 inside_fragoff;
b64c9256 BS	272	int inside_hdrlen;
	273
	274	icmph = skb_header_pointer(skb, outside_hdrlen,
	275	sizeof(_icmph), &_icmph);
	276	if (icmph == NULL)
	277	return 1;
	278
	279	if (icmph->icmp6_type & ICMPV6_INFOMSG_MASK)
	280	return 1;
	281
78296c97 ED	282	inside_iph = skb_header_pointer(skb, outside_hdrlen + sizeof(_icmph),
78296c97 ED	283	sizeof(*ipv6_var), ipv6_var);
b64c9256 BS	284	if (inside_iph == NULL)
	285	return 1;
	286	inside_nexthdr = inside_iph->nexthdr;
	287
78296c97 ED	288	inside_hdrlen = ipv6_skip_exthdr(skb, outside_hdrlen + sizeof(_icmph) +
78296c97 ED	289	sizeof(*ipv6_var),
75f2811c	290	&inside_nexthdr, &inside_fragoff);
b64c9256 BS	291	if (inside_hdrlen < 0)
	292	return 1; /* hjm: Packet has no/incomplete transport layer headers. */
	293
	294	if (inside_nexthdr != IPPROTO_TCP &&
	295	inside_nexthdr != IPPROTO_UDP)
	296	return 1;
	297
	298	ports = skb_header_pointer(skb, inside_hdrlen,
	299	sizeof(_ports), &_ports);
	300	if (ports == NULL)
	301	return 1;
	302
	303	/* the inside IP packet is the one quoted from our side, thus
	304	* its saddr is the local address */
	305	*protocol = inside_nexthdr;
	306	*laddr = &inside_iph->saddr;
	307	*lport = ports[0];
	308	*raddr = &inside_iph->daddr;
	309	*rport = ports[1];
	310
	311	return 0;
	312	}
	313
93742cf8 FW	314	static struct sock *
	315	xt_socket_get_sock_v6(struct net *net, const u8 protocol,
	316	const struct in6_addr saddr, const struct in6_addr daddr,
	317	const __be16 sport, const __be16 dport,
	318	const struct net_device *in)
	319	{
	320	switch (protocol) {
	321	case IPPROTO_TCP:
	322	return inet6_lookup(net, &tcp_hashinfo,
	323	saddr, sport, daddr, dport,
	324	in->ifindex);
	325	case IPPROTO_UDP:
	326	return udp6_lib_lookup(net, saddr, sport, daddr, dport,
	327	in->ifindex);
	328	}
	329
	330	return NULL;
	331	}
	332
d64d80a2 DB	333	static struct sock xt_socket_lookup_slow_v6(const struct sk_buff skb,
d64d80a2 DB	334	const struct net_device *indev)
b64c9256	335	{
6703aa74	336	__be16 uninitialized_var(dport), uninitialized_var(sport);
d64d80a2 DB	337	const struct in6_addr daddr = NULL, saddr = NULL;
	338	struct ipv6hdr *iph = ipv6_hdr(skb);
	339	int thoff = 0, tproto;
b64c9256	340
84018f55	341	tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL);
b64c9256 BS	342	if (tproto < 0) {
b64c9256 BS	343	pr_debug("unable to find transport header in IPv6 packet, dropping\n");
d64d80a2	344	return NULL;
b64c9256 BS	345	}
	346
	347	if (tproto == IPPROTO_UDP \|\| tproto == IPPROTO_TCP) {
d64d80a2 DB	348	struct udphdr _hdr, *hp;
	349
	350	hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr);
b64c9256	351	if (hp == NULL)
d64d80a2	352	return NULL;
b64c9256 BS	353
	354	saddr = &iph->saddr;
	355	sport = hp->source;
	356	daddr = &iph->daddr;
	357	dport = hp->dest;
	358
	359	} else if (tproto == IPPROTO_ICMPV6) {
d64d80a2 DB	360	struct ipv6hdr ipv6_var;
d64d80a2 DB	361
b64c9256	362	if (extract_icmp6_fields(skb, thoff, &tproto, &saddr, &daddr,
78296c97	363	&sport, &dport, &ipv6_var))
d64d80a2	364	return NULL;
b64c9256	365	} else {
d64d80a2	366	return NULL;
b64c9256 BS	367	}
b64c9256 BS	368
d64d80a2 DB	369	return xt_socket_get_sock_v6(dev_net(skb->dev), tproto, saddr, daddr,
	370	sport, dport, indev);
	371	}
	372
	373	static bool
	374	socket_mt6_v1_v2(const struct sk_buff skb, struct xt_action_param par)
	375	{
	376	const struct xt_socket_mtinfo1 info = (struct xt_socket_mtinfo1 ) par->matchinfo;
	377	struct sock *sk = skb->sk;
	378
00028aa3	379	if (!sk)
d64d80a2	380	sk = xt_socket_lookup_slow_v6(skb, par->in);
00028aa3	381	if (sk) {
b64c9256 BS	382	bool wildcard;
	383	bool transparent = true;
	384
681f130f ED	385	/* Ignore sockets listening on INADDR_ANY
	386	* unless XT_SOCKET_NOWILDCARD is set
	387	*/
	388	wildcard = (!(info->flags & XT_SOCKET_NOWILDCARD) &&
a9407000	389	sk_fullsock(sk) &&
efe4208f	390	ipv6_addr_any(&sk->sk_v6_rcv_saddr));
b64c9256 BS	391
b64c9256 BS	392	/* Ignore non-transparent sockets,
a9407000 ED	393	* if XT_SOCKET_TRANSPARENT is used
a9407000 ED	394	*/
baf60efa	395	if (info->flags & XT_SOCKET_TRANSPARENT)
a9407000	396	transparent = xt_socket_sk_is_transparent(sk);
b64c9256	397
00028aa3	398	if (sk != skb->sk)
1a8bf6ee	399	sock_gen_put(sk);
b64c9256 BS	400
	401	if (wildcard \|\| !transparent)
	402	sk = NULL;
	403	}
	404
d64d80a2	405	return sk != NULL;
b64c9256 BS	406	}
	407	#endif
	408
681f130f ED	409	static int socket_mt_v1_check(const struct xt_mtchk_param *par)
	410	{
	411	const struct xt_socket_mtinfo1 info = (struct xt_socket_mtinfo1 ) par->matchinfo;
	412
	413	if (info->flags & ~XT_SOCKET_FLAGS_V1) {
	414	pr_info("unknown flags 0x%x\n", info->flags & ~XT_SOCKET_FLAGS_V1);
	415	return -EINVAL;
	416	}
	417	return 0;
	418	}
	419
	420	static int socket_mt_v2_check(const struct xt_mtchk_param *par)
	421	{
	422	const struct xt_socket_mtinfo2 info = (struct xt_socket_mtinfo2 ) par->matchinfo;
	423
	424	if (info->flags & ~XT_SOCKET_FLAGS_V2) {
	425	pr_info("unknown flags 0x%x\n", info->flags & ~XT_SOCKET_FLAGS_V2);
	426	return -EINVAL;
	427	}
	428	return 0;
	429	}
	430
a31e1ffd LAT	431	static struct xt_match socket_mt_reg[] __read_mostly = {
	432	{
	433	.name = "socket",
	434	.revision = 0,
	435	.family = NFPROTO_IPV4,
b64c9256	436	.match = socket_mt4_v0,
aa3c487f JE	437	.hooks = (1 << NF_INET_PRE_ROUTING) \|
aa3c487f JE	438	(1 << NF_INET_LOCAL_IN),
a31e1ffd LAT	439	.me = THIS_MODULE,
	440	},
	441	{
	442	.name = "socket",
	443	.revision = 1,
	444	.family = NFPROTO_IPV4,
681f130f ED	445	.match = socket_mt4_v1_v2,
681f130f ED	446	.checkentry = socket_mt_v1_check,
a31e1ffd	447	.matchsize = sizeof(struct xt_socket_mtinfo1),
aa3c487f JE	448	.hooks = (1 << NF_INET_PRE_ROUTING) \|
aa3c487f JE	449	(1 << NF_INET_LOCAL_IN),
a31e1ffd LAT	450	.me = THIS_MODULE,
a31e1ffd LAT	451	},
f6318e55	452	#ifdef XT_SOCKET_HAVE_IPV6
b64c9256 BS	453	{
	454	.name = "socket",
	455	.revision = 1,
	456	.family = NFPROTO_IPV6,
681f130f ED	457	.match = socket_mt6_v1_v2,
	458	.checkentry = socket_mt_v1_check,
	459	.matchsize = sizeof(struct xt_socket_mtinfo1),
	460	.hooks = (1 << NF_INET_PRE_ROUTING) \|
	461	(1 << NF_INET_LOCAL_IN),
	462	.me = THIS_MODULE,
	463	},
	464	#endif
	465	{
	466	.name = "socket",
	467	.revision = 2,
	468	.family = NFPROTO_IPV4,
	469	.match = socket_mt4_v1_v2,
	470	.checkentry = socket_mt_v2_check,
	471	.matchsize = sizeof(struct xt_socket_mtinfo1),
	472	.hooks = (1 << NF_INET_PRE_ROUTING) \|
	473	(1 << NF_INET_LOCAL_IN),
	474	.me = THIS_MODULE,
	475	},
	476	#ifdef XT_SOCKET_HAVE_IPV6
	477	{
	478	.name = "socket",
	479	.revision = 2,
	480	.family = NFPROTO_IPV6,
	481	.match = socket_mt6_v1_v2,
	482	.checkentry = socket_mt_v2_check,
b64c9256 BS	483	.matchsize = sizeof(struct xt_socket_mtinfo1),
	484	.hooks = (1 << NF_INET_PRE_ROUTING) \|
	485	(1 << NF_INET_LOCAL_IN),
	486	.me = THIS_MODULE,
	487	},
	488	#endif
136cdc71 KK	489	};
	490
	491	static int __init socket_mt_init(void)
	492	{
	493	nf_defrag_ipv4_enable();
f6318e55	494	#ifdef XT_SOCKET_HAVE_IPV6
b64c9256 BS	495	nf_defrag_ipv6_enable();
	496	#endif
	497
a31e1ffd	498	return xt_register_matches(socket_mt_reg, ARRAY_SIZE(socket_mt_reg));
136cdc71 KK	499	}
	500
	501	static void __exit socket_mt_exit(void)
	502	{
a31e1ffd	503	xt_unregister_matches(socket_mt_reg, ARRAY_SIZE(socket_mt_reg));
136cdc71 KK	504	}
	505
	506	module_init(socket_mt_init);
	507	module_exit(socket_mt_exit);
	508
	509	MODULE_LICENSE("GPL");
	510	MODULE_AUTHOR("Krisztian Kovacs, Balazs Scheidler");
	511	MODULE_DESCRIPTION("x_tables socket match module");
	512	MODULE_ALIAS("ipt_socket");
b64c9256	513	MODULE_ALIAS("ip6t_socket");