[GitHub/mt8127/android_kernel_alcatel_ttab.git] / net / l2tp / l2tp_ip.c

/*
 * L2TPv3 IP encapsulation support
 *
 * Copyright (c) 2008,2009,2010 Katalix Systems Ltd
 *
 *	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/icmp.h>
#include <linux/module.h>
#include <linux/skbuff.h>
#include <linux/random.h>
#include <linux/socket.h>
#include <linux/l2tp.h>
#include <linux/in.h>
#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
#include <net/udp.h>
#include <net/inet_common.h>
#include <net/inet_hashtables.h>
#include <net/tcp_states.h>
#include <net/protocol.h>
#include <net/xfrm.h>

#include "l2tp_core.h"

struct l2tp_ip_sock {
	/* inet_sock has to be the first member of l2tp_ip_sock */
	struct inet_sock	inet;

	u32			conn_id;
	u32			peer_conn_id;
};

static DEFINE_RWLOCK(l2tp_ip_lock);
static struct hlist_head l2tp_ip_table;
static struct hlist_head l2tp_ip_bind_table;

static inline struct l2tp_ip_sock *l2tp_ip_sk(const struct sock *sk)
{
	return (struct l2tp_ip_sock *)sk;
}

static struct sock *__l2tp_ip_bind_lookup(struct net *net, __be32 laddr, int dif, u32 tunnel_id)
{
	struct hlist_node *node;
	struct sock *sk;

	sk_for_each_bound(sk, node, &l2tp_ip_bind_table) {
		struct inet_sock *inet = inet_sk(sk);
		struct l2tp_ip_sock *l2tp = l2tp_ip_sk(sk);

		if (l2tp == NULL)
			continue;

		if ((l2tp->conn_id == tunnel_id) &&
		    net_eq(sock_net(sk), net) &&
		    !(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) &&
		    !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif))
			goto found;
	}

	sk = NULL;
found:
	return sk;
}

static inline struct sock *l2tp_ip_bind_lookup(struct net *net, __be32 laddr, int dif, u32 tunnel_id)
{
	struct sock *sk = __l2tp_ip_bind_lookup(net, laddr, dif, tunnel_id);
	if (sk)
		sock_hold(sk);

	return sk;
}

/* When processing receive frames, there are two cases to
 * consider. Data frames consist of a non-zero session-id and an
 * optional cookie. Control frames consist of a regular L2TP header
 * preceded by 32-bits of zeros.
 *
 * L2TPv3 Session Header Over IP
 *
 *  0                   1                   2                   3
 *  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                           Session ID                          |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |               Cookie (optional, maximum 64 bits)...
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *                                                                 |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *
 * L2TPv3 Control Message Header Over IP
 *
 *  0                   1                   2                   3
 *  0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                      (32 bits of zeros)                       |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |T|L|x|x|S|x|x|x|x|x|x|x|  Ver  |             Length            |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |                     Control Connection ID                     |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 * |               Ns              |               Nr              |
 * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 *
 * All control frames are passed to userspace.
 */
static int l2tp_ip_recv(struct sk_buff *skb)
{
	struct sock *sk;
	u32 session_id;
	u32 tunnel_id;
	unsigned char *ptr, *optr;
	struct l2tp_session *session;
	struct l2tp_tunnel *tunnel = NULL;
	int length;
	int offset;

	/* Point to L2TP header */
	optr = ptr = skb->data;

	if (!pskb_may_pull(skb, 4))
		goto discard;

	session_id = ntohl(*((__be32 *) ptr));
	ptr += 4;

	/* RFC3931: L2TP/IP packets have the first 4 bytes containing
	 * the session_id. If it is 0, the packet is a L2TP control
	 * frame and the session_id value can be discarded.
	 */
	if (session_id == 0) {
		__skb_pull(skb, 4);
		goto pass_up;
	}

	/* Ok, this is a data packet. Lookup the session. */
	session = l2tp_session_find(&init_net, NULL, session_id);
	if (session == NULL)
		goto discard;

	tunnel = session->tunnel;
	if (tunnel == NULL)
		goto discard;

	/* Trace packet contents, if enabled */
	if (tunnel->debug & L2TP_MSG_DATA) {
		length = min(32u, skb->len);
		if (!pskb_may_pull(skb, length))
			goto discard;

		printk(KERN_DEBUG "%s: ip recv: ", tunnel->name);

		offset = 0;
		do {
			printk(" %02X", ptr[offset]);
		} while (++offset < length);

		printk("\n");
	}

	l2tp_recv_common(session, skb, ptr, optr, 0, skb->len, tunnel->recv_payload_hook);

	return 0;

pass_up:
	/* Get the tunnel_id from the L2TP header */
	if (!pskb_may_pull(skb, 12))
		goto discard;

	if ((skb->data[0] & 0xc0) != 0xc0)
		goto discard;

	tunnel_id = ntohl(*(__be32 *) &skb->data[4]);
	tunnel = l2tp_tunnel_find(&init_net, tunnel_id);
	if (tunnel != NULL)
		sk = tunnel->sock;
	else {
		struct iphdr *iph = (struct iphdr *) skb_network_header(skb);

		read_lock_bh(&l2tp_ip_lock);
		sk = __l2tp_ip_bind_lookup(&init_net, iph->daddr, 0, tunnel_id);
		read_unlock_bh(&l2tp_ip_lock);
	}

	if (sk == NULL)
		goto discard;

	sock_hold(sk);

	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
		goto discard_put;

	nf_reset(skb);

	return sk_receive_skb(sk, skb, 1);

discard_put:
	sock_put(sk);

discard:
	kfree_skb(skb);
	return 0;
}

static int l2tp_ip_open(struct sock *sk)
{
	/* Prevent autobind. We don't have ports. */
	inet_sk(sk)->inet_num = IPPROTO_L2TP;

	write_lock_bh(&l2tp_ip_lock);
	sk_add_node(sk, &l2tp_ip_table);
	write_unlock_bh(&l2tp_ip_lock);

	return 0;
}

static void l2tp_ip_close(struct sock *sk, long timeout)
{
	write_lock_bh(&l2tp_ip_lock);
	hlist_del_init(&sk->sk_bind_node);
	sk_del_node_init(sk);
	write_unlock_bh(&l2tp_ip_lock);
	sk_common_release(sk);
}

static void l2tp_ip_destroy_sock(struct sock *sk)
{
	struct sk_buff *skb;

	while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL)
		kfree_skb(skb);

	sk_refcnt_debug_dec(sk);
}

static int l2tp_ip_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct sockaddr_l2tpip *addr = (struct sockaddr_l2tpip *) uaddr;
	int ret = -EINVAL;
	int chk_addr_ret;

	ret = -EADDRINUSE;
	read_lock_bh(&l2tp_ip_lock);
	if (__l2tp_ip_bind_lookup(&init_net, addr->l2tp_addr.s_addr, sk->sk_bound_dev_if, addr->l2tp_conn_id))
		goto out_in_use;

	read_unlock_bh(&l2tp_ip_lock);

	lock_sock(sk);
	if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_l2tpip))
		goto out;

	chk_addr_ret = inet_addr_type(&init_net, addr->l2tp_addr.s_addr);
	ret = -EADDRNOTAVAIL;
	if (addr->l2tp_addr.s_addr && chk_addr_ret != RTN_LOCAL &&
	    chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST)
		goto out;

	if (addr->l2tp_addr.s_addr)
		inet->inet_rcv_saddr = inet->inet_saddr = addr->l2tp_addr.s_addr;
	if (chk_addr_ret == RTN_MULTICAST || chk_addr_ret == RTN_BROADCAST)
		inet->inet_saddr = 0;  /* Use device */
	sk_dst_reset(sk);

	l2tp_ip_sk(sk)->conn_id = addr->l2tp_conn_id;

	write_lock_bh(&l2tp_ip_lock);
	sk_add_bind_node(sk, &l2tp_ip_bind_table);
	sk_del_node_init(sk);
	write_unlock_bh(&l2tp_ip_lock);
	ret = 0;
out:
	release_sock(sk);

	return ret;

out_in_use:
	read_unlock_bh(&l2tp_ip_lock);

	return ret;
}

static int l2tp_ip_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
{
	struct sockaddr_l2tpip *lsa = (struct sockaddr_l2tpip *) uaddr;
	int rc;

	if (addr_len < sizeof(*lsa))
		return -EINVAL;

	if (ipv4_is_multicast(lsa->l2tp_addr.s_addr))
		return -EINVAL;

	rc = ip4_datagram_connect(sk, uaddr, addr_len);
	if (rc < 0)
		return rc;

	lock_sock(sk);

	l2tp_ip_sk(sk)->peer_conn_id = lsa->l2tp_conn_id;

	write_lock_bh(&l2tp_ip_lock);
	hlist_del_init(&sk->sk_bind_node);
	sk_add_bind_node(sk, &l2tp_ip_bind_table);
	write_unlock_bh(&l2tp_ip_lock);

	release_sock(sk);
	return rc;
}

static int l2tp_ip_getname(struct socket *sock, struct sockaddr *uaddr,
			   int *uaddr_len, int peer)
{
	struct sock *sk		= sock->sk;
	struct inet_sock *inet	= inet_sk(sk);
	struct l2tp_ip_sock *lsk = l2tp_ip_sk(sk);
	struct sockaddr_l2tpip *lsa = (struct sockaddr_l2tpip *)uaddr;

	memset(lsa, 0, sizeof(*lsa));
	lsa->l2tp_family = AF_INET;
	if (peer) {
		if (!inet->inet_dport)
			return -ENOTCONN;
		lsa->l2tp_conn_id = lsk->peer_conn_id;
		lsa->l2tp_addr.s_addr = inet->inet_daddr;
	} else {
		__be32 addr = inet->inet_rcv_saddr;
		if (!addr)
			addr = inet->inet_saddr;
		lsa->l2tp_conn_id = lsk->conn_id;
		lsa->l2tp_addr.s_addr = addr;
	}
	*uaddr_len = sizeof(*lsa);
	return 0;
}

static int l2tp_ip_backlog_recv(struct sock *sk, struct sk_buff *skb)
{
	int rc;

	/* Charge it to the socket, dropping if the queue is full. */
	rc = sock_queue_rcv_skb(sk, skb);
	if (rc < 0)
		goto drop;

	return 0;

drop:
	IP_INC_STATS(&init_net, IPSTATS_MIB_INDISCARDS);
	kfree_skb(skb);
	return -1;
}

/* Userspace will call sendmsg() on the tunnel socket to send L2TP
 * control frames.
 */
static int l2tp_ip_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, size_t len)
{
	struct sk_buff *skb;
	int rc;
	struct inet_sock *inet = inet_sk(sk);
	struct rtable *rt = NULL;
	struct flowi4 *fl4;
	int connected = 0;
	__be32 daddr;

	lock_sock(sk);

	rc = -ENOTCONN;
	if (sock_flag(sk, SOCK_DEAD))
		goto out;

	/* Get and verify the address. */
	if (msg->msg_name) {
		struct sockaddr_l2tpip *lip = (struct sockaddr_l2tpip *) msg->msg_name;
		rc = -EINVAL;
		if (msg->msg_namelen < sizeof(*lip))
			goto out;

		if (lip->l2tp_family != AF_INET) {
			rc = -EAFNOSUPPORT;
			if (lip->l2tp_family != AF_UNSPEC)
				goto out;
		}

		daddr = lip->l2tp_addr.s_addr;
	} else {
		rc = -EDESTADDRREQ;
		if (sk->sk_state != TCP_ESTABLISHED)
			goto out;

		daddr = inet->inet_daddr;
		connected = 1;
	}

	/* Allocate a socket buffer */
	rc = -ENOMEM;
	skb = sock_wmalloc(sk, 2 + NET_SKB_PAD + sizeof(struct iphdr) +
			   4 + len, 0, GFP_KERNEL);
	if (!skb)
		goto error;

	/* Reserve space for headers, putting IP header on 4-byte boundary. */
	skb_reserve(skb, 2 + NET_SKB_PAD);
	skb_reset_network_header(skb);
	skb_reserve(skb, sizeof(struct iphdr));
	skb_reset_transport_header(skb);

	/* Insert 0 session_id */
	*((__be32 *) skb_put(skb, 4)) = 0;

	/* Copy user data into skb */
	rc = memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len);
	if (rc < 0) {
		kfree_skb(skb);
		goto error;
	}

	fl4 = &inet->cork.fl.u.ip4;
	if (connected)
		rt = (struct rtable *) __sk_dst_check(sk, 0);

	rcu_read_lock();
	if (rt == NULL) {
		const struct ip_options_rcu *inet_opt;

		inet_opt = rcu_dereference(inet->inet_opt);

		/* Use correct destination address if we have options. */
		if (inet_opt && inet_opt->opt.srr)
			daddr = inet_opt->opt.faddr;

		/* If this fails, retransmit mechanism of transport layer will
		 * keep trying until route appears or the connection times
		 * itself out.
		 */
		rt = ip_route_output_ports(sock_net(sk), fl4, sk,
					   daddr, inet->inet_saddr,
					   inet->inet_dport, inet->inet_sport,
					   sk->sk_protocol, RT_CONN_FLAGS(sk),
					   sk->sk_bound_dev_if);
		if (IS_ERR(rt))
			goto no_route;
		if (connected)
			sk_setup_caps(sk, &rt->dst);
		else
			dst_release(&rt->dst); /* safe since we hold rcu_read_lock */
	}

	/* We dont need to clone dst here, it is guaranteed to not disappear.
	 *  __dev_xmit_skb() might force a refcount if needed.
	 */
	skb_dst_set_noref(skb, &rt->dst);

	/* Queue the packet to IP for output */
	rc = ip_queue_xmit(skb, &inet->cork.fl);
	rcu_read_unlock();

error:
	if (rc >= 0)
		rc = len;

out:
	release_sock(sk);
	return rc;

no_route:
	rcu_read_unlock();
	IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
	kfree_skb(skb);
	rc = -EHOSTUNREACH;
	goto out;
}

static int l2tp_ip_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
			   size_t len, int noblock, int flags, int *addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	size_t copied = 0;
	int err = -EOPNOTSUPP;
	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
	struct sk_buff *skb;

	if (flags & MSG_OOB)
		goto out;

	if (addr_len)
		*addr_len = sizeof(*sin);

	skb = skb_recv_datagram(sk, flags, noblock, &err);
	if (!skb)
		goto out;

	copied = skb->len;
	if (len < copied) {
		msg->msg_flags |= MSG_TRUNC;
		copied = len;
	}

	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
	if (err)
		goto done;

	sock_recv_timestamp(msg, sk, skb);

	/* Copy the address. */
	if (sin) {
		sin->sin_family = AF_INET;
		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
		sin->sin_port = 0;
		memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
	}
	if (inet->cmsg_flags)
		ip_cmsg_recv(msg, skb);
	if (flags & MSG_TRUNC)
		copied = skb->len;
done:
	skb_free_datagram(sk, skb);
out:
	return err ? err : copied;
}

static struct proto l2tp_ip_prot = {
	.name		   = "L2TP/IP",
	.owner		   = THIS_MODULE,
	.init		   = l2tp_ip_open,
	.close		   = l2tp_ip_close,
	.bind		   = l2tp_ip_bind,
	.connect	   = l2tp_ip_connect,
	.disconnect	   = udp_disconnect,
	.ioctl		   = udp_ioctl,
	.destroy	   = l2tp_ip_destroy_sock,
	.setsockopt	   = ip_setsockopt,
	.getsockopt	   = ip_getsockopt,
	.sendmsg	   = l2tp_ip_sendmsg,
	.recvmsg	   = l2tp_ip_recvmsg,
	.backlog_rcv	   = l2tp_ip_backlog_recv,
	.hash		   = inet_hash,
	.unhash		   = inet_unhash,
	.obj_size	   = sizeof(struct l2tp_ip_sock),
#ifdef CONFIG_COMPAT
	.compat_setsockopt = compat_ip_setsockopt,
	.compat_getsockopt = compat_ip_getsockopt,
#endif
};

static const struct proto_ops l2tp_ip_ops = {
	.family		   = PF_INET,
	.owner		   = THIS_MODULE,
	.release	   = inet_release,
	.bind		   = inet_bind,
	.connect	   = inet_dgram_connect,
	.socketpair	   = sock_no_socketpair,
	.accept		   = sock_no_accept,
	.getname	   = l2tp_ip_getname,
	.poll		   = datagram_poll,
	.ioctl		   = inet_ioctl,
	.listen		   = sock_no_listen,
	.shutdown	   = inet_shutdown,
	.setsockopt	   = sock_common_setsockopt,
	.getsockopt	   = sock_common_getsockopt,
	.sendmsg	   = inet_sendmsg,
	.recvmsg	   = sock_common_recvmsg,
	.mmap		   = sock_no_mmap,
	.sendpage	   = sock_no_sendpage,
#ifdef CONFIG_COMPAT
	.compat_setsockopt = compat_sock_common_setsockopt,
	.compat_getsockopt = compat_sock_common_getsockopt,
#endif
};

static struct inet_protosw l2tp_ip_protosw = {
	.type		= SOCK_DGRAM,
	.protocol	= IPPROTO_L2TP,
	.prot		= &l2tp_ip_prot,
	.ops		= &l2tp_ip_ops,
	.no_check	= 0,
};

static struct net_protocol l2tp_ip_protocol __read_mostly = {
	.handler	= l2tp_ip_recv,
};

static int __init l2tp_ip_init(void)
{
	int err;

	printk(KERN_INFO "L2TP IP encapsulation support (L2TPv3)\n");

	err = proto_register(&l2tp_ip_prot, 1);
	if (err != 0)
		goto out;

	err = inet_add_protocol(&l2tp_ip_protocol, IPPROTO_L2TP);
	if (err)
		goto out1;

	inet_register_protosw(&l2tp_ip_protosw);
	return 0;

out1:
	proto_unregister(&l2tp_ip_prot);
out:
	return err;
}

static void __exit l2tp_ip_exit(void)
{
	inet_unregister_protosw(&l2tp_ip_protosw);
	inet_del_protocol(&l2tp_ip_protocol, IPPROTO_L2TP);
	proto_unregister(&l2tp_ip_prot);
}

module_init(l2tp_ip_init);
module_exit(l2tp_ip_exit);

MODULE_LICENSE("GPL");
MODULE_AUTHOR("James Chapman <jchapman@katalix.com>");
MODULE_DESCRIPTION("L2TP over IP");
MODULE_VERSION("1.0");

/* Use the value of SOCK_DGRAM (2) directory, because __stringify doesn't like
 * enums
 */
MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 2, IPPROTO_L2TP);
Commit	Line	Data
0d76751f JC	1	/*
	2	* L2TPv3 IP encapsulation support
	3	*
	4	* Copyright (c) 2008,2009,2010 Katalix Systems Ltd
	5	*
	6	* This program is free software; you can redistribute it and/or
	7	* modify it under the terms of the GNU General Public License
	8	* as published by the Free Software Foundation; either version
	9	* 2 of the License, or (at your option) any later version.
	10	*/
	11
	12	#include <linux/icmp.h>
	13	#include <linux/module.h>
	14	#include <linux/skbuff.h>
	15	#include <linux/random.h>
	16	#include <linux/socket.h>
	17	#include <linux/l2tp.h>
	18	#include <linux/in.h>
	19	#include <net/sock.h>
	20	#include <net/ip.h>
	21	#include <net/icmp.h>
	22	#include <net/udp.h>
	23	#include <net/inet_common.h>
	24	#include <net/inet_hashtables.h>
	25	#include <net/tcp_states.h>
	26	#include <net/protocol.h>
	27	#include <net/xfrm.h>
	28
	29	#include "l2tp_core.h"
	30
	31	struct l2tp_ip_sock {
	32	/* inet_sock has to be the first member of l2tp_ip_sock */
	33	struct inet_sock inet;
	34
c8657fd5 JC	35	u32 conn_id;
c8657fd5 JC	36	u32 peer_conn_id;
0d76751f JC	37	};
	38
	39	static DEFINE_RWLOCK(l2tp_ip_lock);
	40	static struct hlist_head l2tp_ip_table;
	41	static struct hlist_head l2tp_ip_bind_table;
	42
	43	static inline struct l2tp_ip_sock l2tp_ip_sk(const struct sock sk)
	44	{
	45	return (struct l2tp_ip_sock *)sk;
	46	}
	47
	48	static struct sock __l2tp_ip_bind_lookup(struct net net, __be32 laddr, int dif, u32 tunnel_id)
	49	{
	50	struct hlist_node *node;
	51	struct sock *sk;
	52
	53	sk_for_each_bound(sk, node, &l2tp_ip_bind_table) {
	54	struct inet_sock *inet = inet_sk(sk);
	55	struct l2tp_ip_sock *l2tp = l2tp_ip_sk(sk);
	56
	57	if (l2tp == NULL)
	58	continue;
	59
	60	if ((l2tp->conn_id == tunnel_id) &&
e83726b0	61	net_eq(sock_net(sk), net) &&
0d76751f JC	62	!(inet->inet_rcv_saddr && inet->inet_rcv_saddr != laddr) &&
	63	!(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif))
	64	goto found;
	65	}
	66
	67	sk = NULL;
	68	found:
	69	return sk;
	70	}
	71
	72	static inline struct sock l2tp_ip_bind_lookup(struct net net, __be32 laddr, int dif, u32 tunnel_id)
	73	{
	74	struct sock *sk = __l2tp_ip_bind_lookup(net, laddr, dif, tunnel_id);
	75	if (sk)
	76	sock_hold(sk);
	77
	78	return sk;
	79	}
	80
	81	/* When processing receive frames, there are two cases to
	82	* consider. Data frames consist of a non-zero session-id and an
	83	* optional cookie. Control frames consist of a regular L2TP header
	84	* preceded by 32-bits of zeros.
	85	*
	86	* L2TPv3 Session Header Over IP
	87	*
	88	* 0 1 2 3
	89	* 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
	90	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	91	* \| Session ID \|
	92	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	93	* \| Cookie (optional, maximum 64 bits)...
	94	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	95	* \|
	96	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	97	*
	98	* L2TPv3 Control Message Header Over IP
	99	*
	100	* 0 1 2 3
	101	* 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
	102	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	103	* \| (32 bits of zeros) \|
	104	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	105	* \|T\|L\|x\|x\|S\|x\|x\|x\|x\|x\|x\|x\| Ver \| Length \|
	106	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	107	* \| Control Connection ID \|
	108	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	109	* \| Ns \| Nr \|
	110	* +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	111	*
	112	* All control frames are passed to userspace.
	113	*/
	114	static int l2tp_ip_recv(struct sk_buff *skb)
	115	{
	116	struct sock *sk;
	117	u32 session_id;
	118	u32 tunnel_id;
	119	unsigned char ptr, optr;
	120	struct l2tp_session *session;
	121	struct l2tp_tunnel *tunnel = NULL;
	122	int length;
	123	int offset;
	124
	125	/* Point to L2TP header */
126	optr = ptr = skb->data;
127
128	if (!pskb_may_pull(skb, 4))
129	goto discard;
130
131	session_id = ntohl(((__be32 ) ptr));
132	ptr += 4;
133
134	/* RFC3931: L2TP/IP packets have the first 4 bytes containing
135	* the session_id. If it is 0, the packet is a L2TP control
136	* frame and the session_id value can be discarded.
137	*/
138	if (session_id == 0) {
139	__skb_pull(skb, 4);
140	goto pass_up;
141	}
142
143	/* Ok, this is a data packet. Lookup the session. */
144	session = l2tp_session_find(&init_net, NULL, session_id);
145	if (session == NULL)
146	goto discard;
147
148	tunnel = session->tunnel;
149	if (tunnel == NULL)
150	goto discard;
151
152	/* Trace packet contents, if enabled */
153	if (tunnel->debug & L2TP_MSG_DATA) {
154	length = min(32u, skb->len);
155	if (!pskb_may_pull(skb, length))
156	goto discard;
157
158	printk(KERN_DEBUG "%s: ip recv: ", tunnel->name);
159
160	offset = 0;
161	do {
162	printk(" %02X", ptr[offset]);
163	} while (++offset < length);
164
165	printk("\n");
166	}
167
168	l2tp_recv_common(session, skb, ptr, optr, 0, skb->len, tunnel->recv_payload_hook);
169
170	return 0;
171
172	pass_up:
173	/* Get the tunnel_id from the L2TP header */
174	if (!pskb_may_pull(skb, 12))
175	goto discard;
176
177	if ((skb->data[0] & 0xc0) != 0xc0)
178	goto discard;
179
180	tunnel_id = ntohl((__be32 ) &skb->data[4]);
181	tunnel = l2tp_tunnel_find(&init_net, tunnel_id);
182	if (tunnel != NULL)
183	sk = tunnel->sock;
184	else {
185	struct iphdr iph = (struct iphdr ) skb_network_header(skb);
186
187	read_lock_bh(&l2tp_ip_lock);
188	sk = __l2tp_ip_bind_lookup(&init_net, iph->daddr, 0, tunnel_id);
189	read_unlock_bh(&l2tp_ip_lock);
190	}
191
192	if (sk == NULL)
193	goto discard;
194
195	sock_hold(sk);
196
197	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
198	goto discard_put;
199
200	nf_reset(skb);
201
202	return sk_receive_skb(sk, skb, 1);
203
204	discard_put:
205	sock_put(sk);
206
207	discard:
208	kfree_skb(skb);
209	return 0;
210	}
211
212	static int l2tp_ip_open(struct sock *sk)
213	{
214	/* Prevent autobind. We don't have ports. */
215	inet_sk(sk)->inet_num = IPPROTO_L2TP;
216
217	write_lock_bh(&l2tp_ip_lock);
218	sk_add_node(sk, &l2tp_ip_table);
219	write_unlock_bh(&l2tp_ip_lock);
220
221	return 0;
222	}
223
224	static void l2tp_ip_close(struct sock *sk, long timeout)
225	{
226	write_lock_bh(&l2tp_ip_lock);
227	hlist_del_init(&sk->sk_bind_node);
d1f224ae	228	sk_del_node_init(sk);
0d76751f JC	229	write_unlock_bh(&l2tp_ip_lock);
	230	sk_common_release(sk);
	231	}
	232
	233	static void l2tp_ip_destroy_sock(struct sock *sk)
	234	{
	235	struct sk_buff *skb;
	236
	237	while ((skb = __skb_dequeue_tail(&sk->sk_write_queue)) != NULL)
	238	kfree_skb(skb);
	239
	240	sk_refcnt_debug_dec(sk);
	241	}
	242
	243	static int l2tp_ip_bind(struct sock sk, struct sockaddr uaddr, int addr_len)
	244	{
	245	struct inet_sock *inet = inet_sk(sk);
	246	struct sockaddr_l2tpip addr = (struct sockaddr_l2tpip ) uaddr;
	247	int ret = -EINVAL;
	248	int chk_addr_ret;
	249
	250	ret = -EADDRINUSE;
	251	read_lock_bh(&l2tp_ip_lock);
	252	if (__l2tp_ip_bind_lookup(&init_net, addr->l2tp_addr.s_addr, sk->sk_bound_dev_if, addr->l2tp_conn_id))
	253	goto out_in_use;
	254
	255	read_unlock_bh(&l2tp_ip_lock);
	256
	257	lock_sock(sk);
	258	if (sk->sk_state != TCP_CLOSE \|\| addr_len < sizeof(struct sockaddr_l2tpip))
	259	goto out;
	260
	261	chk_addr_ret = inet_addr_type(&init_net, addr->l2tp_addr.s_addr);
	262	ret = -EADDRNOTAVAIL;
	263	if (addr->l2tp_addr.s_addr && chk_addr_ret != RTN_LOCAL &&
	264	chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST)
	265	goto out;
	266
c9be48dc JC	267	if (addr->l2tp_addr.s_addr)
c9be48dc JC	268	inet->inet_rcv_saddr = inet->inet_saddr = addr->l2tp_addr.s_addr;
0d76751f JC	269	if (chk_addr_ret == RTN_MULTICAST \|\| chk_addr_ret == RTN_BROADCAST)
	270	inet->inet_saddr = 0; /* Use device */
	271	sk_dst_reset(sk);
	272
	273	l2tp_ip_sk(sk)->conn_id = addr->l2tp_conn_id;
	274
	275	write_lock_bh(&l2tp_ip_lock);
	276	sk_add_bind_node(sk, &l2tp_ip_bind_table);
	277	sk_del_node_init(sk);
	278	write_unlock_bh(&l2tp_ip_lock);
	279	ret = 0;
	280	out:
	281	release_sock(sk);
	282
	283	return ret;
	284
	285	out_in_use:
	286	read_unlock_bh(&l2tp_ip_lock);
	287
	288	return ret;
	289	}
	290
	291	static int l2tp_ip_connect(struct sock sk, struct sockaddr uaddr, int addr_len)
	292	{
0d76751f	293	struct sockaddr_l2tpip lsa = (struct sockaddr_l2tpip ) uaddr;
de3c7a18	294	int rc;
0d76751f	295
0d76751f	296	if (addr_len < sizeof(*lsa))
de3c7a18	297	return -EINVAL;
2f16270f	298
0d76751f	299	if (ipv4_is_multicast(lsa->l2tp_addr.s_addr))
de3c7a18	300	return -EINVAL;
0d76751f	301
de3c7a18 JC	302	rc = ip4_datagram_connect(sk, uaddr, addr_len);
	303	if (rc < 0)
	304	return rc;
0d76751f	305
de3c7a18	306	lock_sock(sk);
0d76751f JC	307
	308	l2tp_ip_sk(sk)->peer_conn_id = lsa->l2tp_conn_id;
	309
0d76751f JC	310	write_lock_bh(&l2tp_ip_lock);
	311	hlist_del_init(&sk->sk_bind_node);
	312	sk_add_bind_node(sk, &l2tp_ip_bind_table);
	313	write_unlock_bh(&l2tp_ip_lock);
	314
2f16270f	315	release_sock(sk);
0d76751f JC	316	return rc;
	317	}
	318
	319	static int l2tp_ip_getname(struct socket sock, struct sockaddr uaddr,
	320	int *uaddr_len, int peer)
	321	{
	322	struct sock *sk = sock->sk;
	323	struct inet_sock *inet = inet_sk(sk);
	324	struct l2tp_ip_sock *lsk = l2tp_ip_sk(sk);
	325	struct sockaddr_l2tpip lsa = (struct sockaddr_l2tpip )uaddr;
	326
	327	memset(lsa, 0, sizeof(*lsa));
	328	lsa->l2tp_family = AF_INET;
	329	if (peer) {
	330	if (!inet->inet_dport)
	331	return -ENOTCONN;
	332	lsa->l2tp_conn_id = lsk->peer_conn_id;
	333	lsa->l2tp_addr.s_addr = inet->inet_daddr;
	334	} else {
	335	__be32 addr = inet->inet_rcv_saddr;
	336	if (!addr)
	337	addr = inet->inet_saddr;
	338	lsa->l2tp_conn_id = lsk->conn_id;
	339	lsa->l2tp_addr.s_addr = addr;
	340	}
	341	uaddr_len = sizeof(lsa);
	342	return 0;
	343	}
	344
	345	static int l2tp_ip_backlog_recv(struct sock sk, struct sk_buff skb)
	346	{
	347	int rc;
	348
0d76751f JC	349	/* Charge it to the socket, dropping if the queue is full. */
	350	rc = sock_queue_rcv_skb(sk, skb);
	351	if (rc < 0)
	352	goto drop;
	353
	354	return 0;
	355
	356	drop:
	357	IP_INC_STATS(&init_net, IPSTATS_MIB_INDISCARDS);
	358	kfree_skb(skb);
	359	return -1;
	360	}
	361
	362	/* Userspace will call sendmsg() on the tunnel socket to send L2TP
	363	* control frames.
	364	*/
	365	static int l2tp_ip_sendmsg(struct kiocb iocb, struct sock sk, struct msghdr *msg, size_t len)
	366	{
	367	struct sk_buff *skb;
	368	int rc;
0d76751f	369	struct inet_sock *inet = inet_sk(sk);
0d76751f	370	struct rtable *rt = NULL;
fdbb0f07	371	struct flowi4 *fl4;
0d76751f JC	372	int connected = 0;
	373	__be32 daddr;
	374
2f16270f DM	375	lock_sock(sk);
	376
	377	rc = -ENOTCONN;
0d76751f	378	if (sock_flag(sk, SOCK_DEAD))
2f16270f	379	goto out;
0d76751f JC	380
	381	/* Get and verify the address. */
	382	if (msg->msg_name) {
	383	struct sockaddr_l2tpip lip = (struct sockaddr_l2tpip ) msg->msg_name;
2f16270f	384	rc = -EINVAL;
0d76751f	385	if (msg->msg_namelen < sizeof(*lip))
2f16270f	386	goto out;
0d76751f JC	387
0d76751f JC	388	if (lip->l2tp_family != AF_INET) {
2f16270f	389	rc = -EAFNOSUPPORT;
0d76751f	390	if (lip->l2tp_family != AF_UNSPEC)
2f16270f	391	goto out;
0d76751f JC	392	}
	393
	394	daddr = lip->l2tp_addr.s_addr;
	395	} else {
84768edb	396	rc = -EDESTADDRREQ;
0d76751f	397	if (sk->sk_state != TCP_ESTABLISHED)
84768edb	398	goto out;
0d76751f JC	399
	400	daddr = inet->inet_daddr;
	401	connected = 1;
	402	}
	403
	404	/* Allocate a socket buffer */
	405	rc = -ENOMEM;
	406	skb = sock_wmalloc(sk, 2 + NET_SKB_PAD + sizeof(struct iphdr) +
	407	4 + len, 0, GFP_KERNEL);
	408	if (!skb)
	409	goto error;
	410
	411	/* Reserve space for headers, putting IP header on 4-byte boundary. */
	412	skb_reserve(skb, 2 + NET_SKB_PAD);
	413	skb_reset_network_header(skb);
	414	skb_reserve(skb, sizeof(struct iphdr));
	415	skb_reset_transport_header(skb);
	416
	417	/* Insert 0 session_id */
	418	((__be32 ) skb_put(skb, 4)) = 0;
	419
	420	/* Copy user data into skb */
	421	rc = memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len);
	422	if (rc < 0) {
	423	kfree_skb(skb);
	424	goto error;
	425	}
	426
fdbb0f07	427	fl4 = &inet->cork.fl.u.ip4;
0d76751f JC	428	if (connected)
	429	rt = (struct rtable *) __sk_dst_check(sk, 0);
	430
081b1b1b	431	rcu_read_lock();
0d76751f	432	if (rt == NULL) {
081b1b1b	433	const struct ip_options_rcu *inet_opt;
f6d8bd05	434
778865a5	435	inet_opt = rcu_dereference(inet->inet_opt);
f6d8bd05	436
0d76751f	437	/* Use correct destination address if we have options. */
f6d8bd05 ED	438	if (inet_opt && inet_opt->opt.srr)
f6d8bd05 ED	439	daddr = inet_opt->opt.faddr;
0d76751f	440
78fbfd8a DM	441	/* If this fails, retransmit mechanism of transport layer will
	442	* keep trying until route appears or the connection times
	443	* itself out.
	444	*/
fdbb0f07	445	rt = ip_route_output_ports(sock_net(sk), fl4, sk,
78fbfd8a DM	446	daddr, inet->inet_saddr,
	447	inet->inet_dport, inet->inet_sport,
	448	sk->sk_protocol, RT_CONN_FLAGS(sk),
	449	sk->sk_bound_dev_if);
	450	if (IS_ERR(rt))
	451	goto no_route;
081b1b1b ED	452	if (connected)
	453	sk_setup_caps(sk, &rt->dst);
	454	else
	455	dst_release(&rt->dst); /* safe since we hold rcu_read_lock */
0d76751f	456	}
081b1b1b ED	457
	458	/* We dont need to clone dst here, it is guaranteed to not disappear.
	459	* __dev_xmit_skb() might force a refcount if needed.
	460	*/
	461	skb_dst_set_noref(skb, &rt->dst);
0d76751f JC	462
0d76751f JC	463	/* Queue the packet to IP for output */
d9d8da80	464	rc = ip_queue_xmit(skb, &inet->cork.fl);
081b1b1b	465	rcu_read_unlock();
0d76751f JC	466
0d76751f JC	467	error:
c8657fd5	468	if (rc >= 0)
0d76751f	469	rc = len;
0d76751f	470
2f16270f DM	471	out:
2f16270f DM	472	release_sock(sk);
0d76751f JC	473	return rc;
	474
	475	no_route:
081b1b1b	476	rcu_read_unlock();
0d76751f JC	477	IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTNOROUTES);
0d76751f JC	478	kfree_skb(skb);
2f16270f DM	479	rc = -EHOSTUNREACH;
2f16270f DM	480	goto out;
0d76751f JC	481	}
	482
	483	static int l2tp_ip_recvmsg(struct kiocb iocb, struct sock sk, struct msghdr *msg,
	484	size_t len, int noblock, int flags, int *addr_len)
	485	{
	486	struct inet_sock *inet = inet_sk(sk);
0d76751f JC	487	size_t copied = 0;
	488	int err = -EOPNOTSUPP;
	489	struct sockaddr_in sin = (struct sockaddr_in )msg->msg_name;
	490	struct sk_buff *skb;
	491
	492	if (flags & MSG_OOB)
	493	goto out;
	494
	495	if (addr_len)
	496	addr_len = sizeof(sin);
	497
	498	skb = skb_recv_datagram(sk, flags, noblock, &err);
	499	if (!skb)
	500	goto out;
	501
	502	copied = skb->len;
	503	if (len < copied) {
	504	msg->msg_flags \|= MSG_TRUNC;
	505	copied = len;
	506	}
	507
	508	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
	509	if (err)
	510	goto done;
	511
	512	sock_recv_timestamp(msg, sk, skb);
	513
	514	/* Copy the address. */
	515	if (sin) {
	516	sin->sin_family = AF_INET;
	517	sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
	518	sin->sin_port = 0;
	519	memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
	520	}
	521	if (inet->cmsg_flags)
	522	ip_cmsg_recv(msg, skb);
	523	if (flags & MSG_TRUNC)
	524	copied = skb->len;
	525	done:
	526	skb_free_datagram(sk, skb);
	527	out:
c8657fd5	528	return err ? err : copied;
0d76751f JC	529	}
0d76751f JC	530
fc130840	531	static struct proto l2tp_ip_prot = {
0d76751f JC	532	.name = "L2TP/IP",
	533	.owner = THIS_MODULE,
	534	.init = l2tp_ip_open,
	535	.close = l2tp_ip_close,
	536	.bind = l2tp_ip_bind,
	537	.connect = l2tp_ip_connect,
	538	.disconnect = udp_disconnect,
	539	.ioctl = udp_ioctl,
	540	.destroy = l2tp_ip_destroy_sock,
	541	.setsockopt = ip_setsockopt,
	542	.getsockopt = ip_getsockopt,
	543	.sendmsg = l2tp_ip_sendmsg,
	544	.recvmsg = l2tp_ip_recvmsg,
	545	.backlog_rcv = l2tp_ip_backlog_recv,
	546	.hash = inet_hash,
	547	.unhash = inet_unhash,
	548	.obj_size = sizeof(struct l2tp_ip_sock),
	549	#ifdef CONFIG_COMPAT
	550	.compat_setsockopt = compat_ip_setsockopt,
	551	.compat_getsockopt = compat_ip_getsockopt,
	552	#endif
	553	};
	554
	555	static const struct proto_ops l2tp_ip_ops = {
	556	.family = PF_INET,
	557	.owner = THIS_MODULE,
	558	.release = inet_release,
	559	.bind = inet_bind,
	560	.connect = inet_dgram_connect,
	561	.socketpair = sock_no_socketpair,
	562	.accept = sock_no_accept,
	563	.getname = l2tp_ip_getname,
	564	.poll = datagram_poll,
	565	.ioctl = inet_ioctl,
	566	.listen = sock_no_listen,
	567	.shutdown = inet_shutdown,
	568	.setsockopt = sock_common_setsockopt,
	569	.getsockopt = sock_common_getsockopt,
	570	.sendmsg = inet_sendmsg,
	571	.recvmsg = sock_common_recvmsg,
	572	.mmap = sock_no_mmap,
	573	.sendpage = sock_no_sendpage,
	574	#ifdef CONFIG_COMPAT
	575	.compat_setsockopt = compat_sock_common_setsockopt,
	576	.compat_getsockopt = compat_sock_common_getsockopt,
	577	#endif
	578	};
	579
	580	static struct inet_protosw l2tp_ip_protosw = {
	581	.type = SOCK_DGRAM,
	582	.protocol = IPPROTO_L2TP,
	583	.prot = &l2tp_ip_prot,
	584	.ops = &l2tp_ip_ops,
	585	.no_check = 0,
	586	};
	587
	588	static struct net_protocol l2tp_ip_protocol __read_mostly = {
	589	.handler = l2tp_ip_recv,
	590	};
	591
	592	static int __init l2tp_ip_init(void)
	593	{
	594	int err;
	595
596	printk(KERN_INFO "L2TP IP encapsulation support (L2TPv3)\n");
597
598	err = proto_register(&l2tp_ip_prot, 1);
599	if (err != 0)
600	goto out;
601
602	err = inet_add_protocol(&l2tp_ip_protocol, IPPROTO_L2TP);
603	if (err)
604	goto out1;
605
606	inet_register_protosw(&l2tp_ip_protosw);
607	return 0;
608
609	out1:
610	proto_unregister(&l2tp_ip_prot);
611	out:
612	return err;
613	}
614
615	static void __exit l2tp_ip_exit(void)
616	{
617	inet_unregister_protosw(&l2tp_ip_protosw);
618	inet_del_protocol(&l2tp_ip_protocol, IPPROTO_L2TP);
619	proto_unregister(&l2tp_ip_prot);
620	}
621
622	module_init(l2tp_ip_init);
623	module_exit(l2tp_ip_exit);
624
625	MODULE_LICENSE("GPL");
626	MODULE_AUTHOR("James Chapman <jchapman@katalix.com>");
627	MODULE_DESCRIPTION("L2TP over IP");
628	MODULE_VERSION("1.0");
e8d34a88	629
e9c54999	630	/* Use the value of SOCK_DGRAM (2) directory, because __stringify doesn't like
e8d34a88 MM	631	* enums
	632	*/
	633	MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 2, IPPROTO_L2TP);