/*
 * net/key/af_key.c	An implementation of PF_KEYv2 sockets.
 *
 *		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.
 *
 * Authors:	Maxim Giryaev	<gem@asplinux.ru>
 *		David S. Miller	<davem@redhat.com>
 *		Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
 *		Kunihiro Ishiguro <kunihiro@ipinfusion.com>
 *		Kazunori MIYAZAWA / USAGI Project <miyazawa@linux-ipv6.org>
 *		Derek Atkins <derek@ihtfp.com>
 */

#include <linux/capability.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/socket.h>
#include <linux/pfkeyv2.h>
#include <linux/ipsec.h>
#include <linux/skbuff.h>
#include <linux/rtnetlink.h>
#include <linux/in.h>
#include <linux/in6.h>
#include <linux/proc_fs.h>
#include <linux/init.h>
#include <net/net_namespace.h>
#include <net/netns/generic.h>
#include <net/xfrm.h>

#include <net/sock.h>

#define _X2KEY(x) ((x) == XFRM_INF ? 0 : (x))
#define _KEY2X(x) ((x) == 0 ? XFRM_INF : (x))

static int pfkey_net_id;
struct netns_pfkey {
	/* List of all pfkey sockets. */
	struct hlist_head table;
	atomic_t socks_nr;
};
static DECLARE_WAIT_QUEUE_HEAD(pfkey_table_wait);
static DEFINE_RWLOCK(pfkey_table_lock);
static atomic_t pfkey_table_users = ATOMIC_INIT(0);

struct pfkey_sock {
	/* struct sock must be the first member of struct pfkey_sock */
	struct sock	sk;
	int		registered;
	int		promisc;

	struct {
		uint8_t		msg_version;
		uint32_t	msg_pid;
		int		(*dump)(struct pfkey_sock *sk);
		void		(*done)(struct pfkey_sock *sk);
		union {
			struct xfrm_policy_walk	policy;
			struct xfrm_state_walk	state;
		} u;
		struct sk_buff	*skb;
	} dump;
};

static inline struct pfkey_sock *pfkey_sk(struct sock *sk)
{
	return (struct pfkey_sock *)sk;
}

static int pfkey_can_dump(struct sock *sk)
{
	if (3 * atomic_read(&sk->sk_rmem_alloc) <= 2 * sk->sk_rcvbuf)
		return 1;
	return 0;
}

static void pfkey_terminate_dump(struct pfkey_sock *pfk)
{
	if (pfk->dump.dump) {
		if (pfk->dump.skb) {
			kfree_skb(pfk->dump.skb);
			pfk->dump.skb = NULL;
		}
		pfk->dump.done(pfk);
		pfk->dump.dump = NULL;
		pfk->dump.done = NULL;
	}
}

static void pfkey_sock_destruct(struct sock *sk)
{
	struct net *net = sock_net(sk);
	struct netns_pfkey *net_pfkey = net_generic(net, pfkey_net_id);

	pfkey_terminate_dump(pfkey_sk(sk));
	skb_queue_purge(&sk->sk_receive_queue);

	if (!sock_flag(sk, SOCK_DEAD)) {
		printk("Attempt to release alive pfkey socket: %p\n", sk);
		return;
	}

	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
	WARN_ON(atomic_read(&sk->sk_wmem_alloc));

	atomic_dec(&net_pfkey->socks_nr);
}

static void pfkey_table_grab(void)
{
	write_lock_bh(&pfkey_table_lock);

	if (atomic_read(&pfkey_table_users)) {
		DECLARE_WAITQUEUE(wait, current);

		add_wait_queue_exclusive(&pfkey_table_wait, &wait);
		for(;;) {
			set_current_state(TASK_UNINTERRUPTIBLE);
			if (atomic_read(&pfkey_table_users) == 0)
				break;
			write_unlock_bh(&pfkey_table_lock);
			schedule();
			write_lock_bh(&pfkey_table_lock);
		}

		__set_current_state(TASK_RUNNING);
		remove_wait_queue(&pfkey_table_wait, &wait);
	}
}

static __inline__ void pfkey_table_ungrab(void)
{
	write_unlock_bh(&pfkey_table_lock);
	wake_up(&pfkey_table_wait);
}

static __inline__ void pfkey_lock_table(void)
{
	/* read_lock() synchronizes us to pfkey_table_grab */

	read_lock(&pfkey_table_lock);
	atomic_inc(&pfkey_table_users);
	read_unlock(&pfkey_table_lock);
}

static __inline__ void pfkey_unlock_table(void)
{
	if (atomic_dec_and_test(&pfkey_table_users))
		wake_up(&pfkey_table_wait);
}


static const struct proto_ops pfkey_ops;

static void pfkey_insert(struct sock *sk)
{
	struct net *net = sock_net(sk);
	struct netns_pfkey *net_pfkey = net_generic(net, pfkey_net_id);

	pfkey_table_grab();
	sk_add_node(sk, &net_pfkey->table);
	pfkey_table_ungrab();
}

static void pfkey_remove(struct sock *sk)
{
	pfkey_table_grab();
	sk_del_node_init(sk);
	pfkey_table_ungrab();
}

static struct proto key_proto = {
	.name	  = "KEY",
	.owner	  = THIS_MODULE,
	.obj_size = sizeof(struct pfkey_sock),
};

static int pfkey_create(struct net *net, struct socket *sock, int protocol)
{
	struct netns_pfkey *net_pfkey = net_generic(net, pfkey_net_id);
	struct sock *sk;
	int err;

	if (!capable(CAP_NET_ADMIN))
		return -EPERM;
	if (sock->type != SOCK_RAW)
		return -ESOCKTNOSUPPORT;
	if (protocol != PF_KEY_V2)
		return -EPROTONOSUPPORT;

	err = -ENOMEM;
	sk = sk_alloc(net, PF_KEY, GFP_KERNEL, &key_proto);
	if (sk == NULL)
		goto out;

	sock->ops = &pfkey_ops;
	sock_init_data(sock, sk);

	sk->sk_family = PF_KEY;
	sk->sk_destruct = pfkey_sock_destruct;

	atomic_inc(&net_pfkey->socks_nr);

	pfkey_insert(sk);

	return 0;
out:
	return err;
}

static int pfkey_release(struct socket *sock)
{
	struct sock *sk = sock->sk;

	if (!sk)
		return 0;

	pfkey_remove(sk);

	sock_orphan(sk);
	sock->sk = NULL;
	skb_queue_purge(&sk->sk_write_queue);
	sock_put(sk);

	return 0;
}

static int pfkey_broadcast_one(struct sk_buff *skb, struct sk_buff **skb2,
			       gfp_t allocation, struct sock *sk)
{
	int err = -ENOBUFS;

	sock_hold(sk);
	if (*skb2 == NULL) {
		if (atomic_read(&skb->users) != 1) {
			*skb2 = skb_clone(skb, allocation);
		} else {
			*skb2 = skb;
			atomic_inc(&skb->users);
		}
	}
	if (*skb2 != NULL) {
		if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf) {
			skb_orphan(*skb2);
			skb_set_owner_r(*skb2, sk);
			skb_queue_tail(&sk->sk_receive_queue, *skb2);
			sk->sk_data_ready(sk, (*skb2)->len);
			*skb2 = NULL;
			err = 0;
		}
	}
	sock_put(sk);
	return err;
}

/* Send SKB to all pfkey sockets matching selected criteria.  */
#define BROADCAST_ALL		0
#define BROADCAST_ONE		1
#define BROADCAST_REGISTERED	2
#define BROADCAST_PROMISC_ONLY	4
static int pfkey_broadcast(struct sk_buff *skb, gfp_t allocation,
			   int broadcast_flags, struct sock *one_sk,
			   struct net *net)
{
	struct netns_pfkey *net_pfkey = net_generic(net, pfkey_net_id);
	struct sock *sk;
	struct hlist_node *node;
	struct sk_buff *skb2 = NULL;
	int err = -ESRCH;

	/* XXX Do we need something like netlink_overrun?  I think
	 * XXX PF_KEY socket apps will not mind current behavior.
	 */
	if (!skb)
		return -ENOMEM;

	pfkey_lock_table();
	sk_for_each(sk, node, &net_pfkey->table) {
		struct pfkey_sock *pfk = pfkey_sk(sk);
		int err2;

		/* Yes, it means that if you are meant to receive this
		 * pfkey message you receive it twice as promiscuous
		 * socket.
		 */
		if (pfk->promisc)
			pfkey_broadcast_one(skb, &skb2, allocation, sk);

		/* the exact target will be processed later */
		if (sk == one_sk)
			continue;
		if (broadcast_flags != BROADCAST_ALL) {
			if (broadcast_flags & BROADCAST_PROMISC_ONLY)
				continue;
			if ((broadcast_flags & BROADCAST_REGISTERED) &&
			    !pfk->registered)
				continue;
			if (broadcast_flags & BROADCAST_ONE)
				continue;
		}

		err2 = pfkey_broadcast_one(skb, &skb2, allocation, sk);

		/* Error is cleare after succecful sending to at least one
		 * registered KM */
		if ((broadcast_flags & BROADCAST_REGISTERED) && err)
			err = err2;
	}
	pfkey_unlock_table();

	if (one_sk != NULL)
		err = pfkey_broadcast_one(skb, &skb2, allocation, one_sk);

	kfree_skb(skb2);
	kfree_skb(skb);
	return err;
}

static int pfkey_do_dump(struct pfkey_sock *pfk)
{
	struct sadb_msg *hdr;
	int rc;

	rc = pfk->dump.dump(pfk);
	if (rc == -ENOBUFS)
		return 0;

	if (pfk->dump.skb) {
		if (!pfkey_can_dump(&pfk->sk))
			return 0;

		hdr = (struct sadb_msg *) pfk->dump.skb->data;
		hdr->sadb_msg_seq = 0;
		hdr->sadb_msg_errno = rc;
		pfkey_broadcast(pfk->dump.skb, GFP_ATOMIC, BROADCAST_ONE,
				&pfk->sk, sock_net(&pfk->sk));
		pfk->dump.skb = NULL;
	}

	pfkey_terminate_dump(pfk);
	return rc;
}

static inline void pfkey_hdr_dup(struct sadb_msg *new, struct sadb_msg *orig)
{
	*new = *orig;
}

static int pfkey_error(struct sadb_msg *orig, int err, struct sock *sk)
{
	struct sk_buff *skb = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_KERNEL);
	struct sadb_msg *hdr;

	if (!skb)
		return -ENOBUFS;

	/* Woe be to the platform trying to support PFKEY yet
	 * having normal errnos outside the 1-255 range, inclusive.
	 */
	err = -err;
	if (err == ERESTARTSYS ||
	    err == ERESTARTNOHAND ||
	    err == ERESTARTNOINTR)
		err = EINTR;
	if (err >= 512)
		err = EINVAL;
	BUG_ON(err <= 0 || err >= 256);

	hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg));
	pfkey_hdr_dup(hdr, orig);
	hdr->sadb_msg_errno = (uint8_t) err;
	hdr->sadb_msg_len = (sizeof(struct sadb_msg) /
			     sizeof(uint64_t));

	pfkey_broadcast(skb, GFP_KERNEL, BROADCAST_ONE, sk, sock_net(sk));

	return 0;
}

static u8 sadb_ext_min_len[] = {
	[SADB_EXT_RESERVED]		= (u8) 0,
	[SADB_EXT_SA]			= (u8) sizeof(struct sadb_sa),
	[SADB_EXT_LIFETIME_CURRENT]	= (u8) sizeof(struct sadb_lifetime),
	[SADB_EXT_LIFETIME_HARD]	= (u8) sizeof(struct sadb_lifetime),
	[SADB_EXT_LIFETIME_SOFT]	= (u8) sizeof(struct sadb_lifetime),
	[SADB_EXT_ADDRESS_SRC]		= (u8) sizeof(struct sadb_address),
	[SADB_EXT_ADDRESS_DST]		= (u8) sizeof(struct sadb_address),
	[SADB_EXT_ADDRESS_PROXY]	= (u8) sizeof(struct sadb_address),
	[SADB_EXT_KEY_AUTH]		= (u8) sizeof(struct sadb_key),
	[SADB_EXT_KEY_ENCRYPT]		= (u8) sizeof(struct sadb_key),
	[SADB_EXT_IDENTITY_SRC]		= (u8) sizeof(struct sadb_ident),
	[SADB_EXT_IDENTITY_DST]		= (u8) sizeof(struct sadb_ident),
	[SADB_EXT_SENSITIVITY]		= (u8) sizeof(struct sadb_sens),
	[SADB_EXT_PROPOSAL]		= (u8) sizeof(struct sadb_prop),
	[SADB_EXT_SUPPORTED_AUTH]	= (u8) sizeof(struct sadb_supported),
	[SADB_EXT_SUPPORTED_ENCRYPT]	= (u8) sizeof(struct sadb_supported),
	[SADB_EXT_SPIRANGE]		= (u8) sizeof(struct sadb_spirange),
	[SADB_X_EXT_KMPRIVATE]		= (u8) sizeof(struct sadb_x_kmprivate),
	[SADB_X_EXT_POLICY]		= (u8) sizeof(struct sadb_x_policy),
	[SADB_X_EXT_SA2]		= (u8) sizeof(struct sadb_x_sa2),
	[SADB_X_EXT_NAT_T_TYPE]		= (u8) sizeof(struct sadb_x_nat_t_type),
	[SADB_X_EXT_NAT_T_SPORT]	= (u8) sizeof(struct sadb_x_nat_t_port),
	[SADB_X_EXT_NAT_T_DPORT]	= (u8) sizeof(struct sadb_x_nat_t_port),
	[SADB_X_EXT_NAT_T_OA]		= (u8) sizeof(struct sadb_address),
	[SADB_X_EXT_SEC_CTX]		= (u8) sizeof(struct sadb_x_sec_ctx),
	[SADB_X_EXT_KMADDRESS]		= (u8) sizeof(struct sadb_x_kmaddress),
};

/* Verify sadb_address_{len,prefixlen} against sa_family.  */
static int verify_address_len(void *p)
{
	struct sadb_address *sp = p;
	struct sockaddr *addr = (struct sockaddr *)(sp + 1);
	struct sockaddr_in *sin;
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	struct sockaddr_in6 *sin6;
#endif
	int len;

	switch (addr->sa_family) {
	case AF_INET:
		len = DIV_ROUND_UP(sizeof(*sp) + sizeof(*sin), sizeof(uint64_t));
		if (sp->sadb_address_len != len ||
		    sp->sadb_address_prefixlen > 32)
			return -EINVAL;
		break;
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	case AF_INET6:
		len = DIV_ROUND_UP(sizeof(*sp) + sizeof(*sin6), sizeof(uint64_t));
		if (sp->sadb_address_len != len ||
		    sp->sadb_address_prefixlen > 128)
			return -EINVAL;
		break;
#endif
	default:
		/* It is user using kernel to keep track of security
		 * associations for another protocol, such as
		 * OSPF/RSVP/RIPV2/MIP.  It is user's job to verify
		 * lengths.
		 *
		 * XXX Actually, association/policy database is not yet
		 * XXX able to cope with arbitrary sockaddr families.
		 * XXX When it can, remove this -EINVAL.  -DaveM
		 */
		return -EINVAL;
		break;
	}

	return 0;
}

static inline int pfkey_sec_ctx_len(struct sadb_x_sec_ctx *sec_ctx)
{
	return DIV_ROUND_UP(sizeof(struct sadb_x_sec_ctx) +
			    sec_ctx->sadb_x_ctx_len,
			    sizeof(uint64_t));
}

static inline int verify_sec_ctx_len(void *p)
{
	struct sadb_x_sec_ctx *sec_ctx = (struct sadb_x_sec_ctx *)p;
	int len = sec_ctx->sadb_x_ctx_len;

	if (len > PAGE_SIZE)
		return -EINVAL;

	len = pfkey_sec_ctx_len(sec_ctx);

	if (sec_ctx->sadb_x_sec_len != len)
		return -EINVAL;

	return 0;
}

static inline struct xfrm_user_sec_ctx *pfkey_sadb2xfrm_user_sec_ctx(struct sadb_x_sec_ctx *sec_ctx)
{
	struct xfrm_user_sec_ctx *uctx = NULL;
	int ctx_size = sec_ctx->sadb_x_ctx_len;

	uctx = kmalloc((sizeof(*uctx)+ctx_size), GFP_KERNEL);

	if (!uctx)
		return NULL;

	uctx->len = pfkey_sec_ctx_len(sec_ctx);
	uctx->exttype = sec_ctx->sadb_x_sec_exttype;
	uctx->ctx_doi = sec_ctx->sadb_x_ctx_doi;
	uctx->ctx_alg = sec_ctx->sadb_x_ctx_alg;
	uctx->ctx_len = sec_ctx->sadb_x_ctx_len;
	memcpy(uctx + 1, sec_ctx + 1,
	       uctx->ctx_len);

	return uctx;
}

static int present_and_same_family(struct sadb_address *src,
				   struct sadb_address *dst)
{
	struct sockaddr *s_addr, *d_addr;

	if (!src || !dst)
		return 0;

	s_addr = (struct sockaddr *)(src + 1);
	d_addr = (struct sockaddr *)(dst + 1);
	if (s_addr->sa_family != d_addr->sa_family)
		return 0;
	if (s_addr->sa_family != AF_INET
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	    && s_addr->sa_family != AF_INET6
#endif
		)
		return 0;

	return 1;
}

static int parse_exthdrs(struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	char *p = (char *) hdr;
	int len = skb->len;

	len -= sizeof(*hdr);
	p += sizeof(*hdr);
	while (len > 0) {
		struct sadb_ext *ehdr = (struct sadb_ext *) p;
		uint16_t ext_type;
		int ext_len;

		ext_len  = ehdr->sadb_ext_len;
		ext_len *= sizeof(uint64_t);
		ext_type = ehdr->sadb_ext_type;
		if (ext_len < sizeof(uint64_t) ||
		    ext_len > len ||
		    ext_type == SADB_EXT_RESERVED)
			return -EINVAL;

		if (ext_type <= SADB_EXT_MAX) {
			int min = (int) sadb_ext_min_len[ext_type];
			if (ext_len < min)
				return -EINVAL;
			if (ext_hdrs[ext_type-1] != NULL)
				return -EINVAL;
			if (ext_type == SADB_EXT_ADDRESS_SRC ||
			    ext_type == SADB_EXT_ADDRESS_DST ||
			    ext_type == SADB_EXT_ADDRESS_PROXY ||
			    ext_type == SADB_X_EXT_NAT_T_OA) {
				if (verify_address_len(p))
					return -EINVAL;
			}
			if (ext_type == SADB_X_EXT_SEC_CTX) {
				if (verify_sec_ctx_len(p))
					return -EINVAL;
			}
			ext_hdrs[ext_type-1] = p;
		}
		p   += ext_len;
		len -= ext_len;
	}

	return 0;
}

static uint16_t
pfkey_satype2proto(uint8_t satype)
{
	switch (satype) {
	case SADB_SATYPE_UNSPEC:
		return IPSEC_PROTO_ANY;
	case SADB_SATYPE_AH:
		return IPPROTO_AH;
	case SADB_SATYPE_ESP:
		return IPPROTO_ESP;
	case SADB_X_SATYPE_IPCOMP:
		return IPPROTO_COMP;
		break;
	default:
		return 0;
	}
	/* NOTREACHED */
}

static uint8_t
pfkey_proto2satype(uint16_t proto)
{
	switch (proto) {
	case IPPROTO_AH:
		return SADB_SATYPE_AH;
	case IPPROTO_ESP:
		return SADB_SATYPE_ESP;
	case IPPROTO_COMP:
		return SADB_X_SATYPE_IPCOMP;
		break;
	default:
		return 0;
	}
	/* NOTREACHED */
}

/* BTW, this scheme means that there is no way with PFKEY2 sockets to
 * say specifically 'just raw sockets' as we encode them as 255.
 */

static uint8_t pfkey_proto_to_xfrm(uint8_t proto)
{
	return (proto == IPSEC_PROTO_ANY ? 0 : proto);
}

static uint8_t pfkey_proto_from_xfrm(uint8_t proto)
{
	return (proto ? proto : IPSEC_PROTO_ANY);
}

static inline int pfkey_sockaddr_len(sa_family_t family)
{
	switch (family) {
	case AF_INET:
		return sizeof(struct sockaddr_in);
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	case AF_INET6:
		return sizeof(struct sockaddr_in6);
#endif
	}
	return 0;
}

static
int pfkey_sockaddr_extract(const struct sockaddr *sa, xfrm_address_t *xaddr)
{
	switch (sa->sa_family) {
	case AF_INET:
		xaddr->a4 =
			((struct sockaddr_in *)sa)->sin_addr.s_addr;
		return AF_INET;
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	case AF_INET6:
		memcpy(xaddr->a6,
		       &((struct sockaddr_in6 *)sa)->sin6_addr,
		       sizeof(struct in6_addr));
		return AF_INET6;
#endif
	}
	return 0;
}

static
int pfkey_sadb_addr2xfrm_addr(struct sadb_address *addr, xfrm_address_t *xaddr)
{
	return pfkey_sockaddr_extract((struct sockaddr *)(addr + 1),
				      xaddr);
}

static struct  xfrm_state *pfkey_xfrm_state_lookup(struct net *net, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct sadb_sa *sa;
	struct sadb_address *addr;
	uint16_t proto;
	unsigned short family;
	xfrm_address_t *xaddr;

	sa = (struct sadb_sa *) ext_hdrs[SADB_EXT_SA-1];
	if (sa == NULL)
		return NULL;

	proto = pfkey_satype2proto(hdr->sadb_msg_satype);
	if (proto == 0)
		return NULL;

	/* sadb_address_len should be checked by caller */
	addr = (struct sadb_address *) ext_hdrs[SADB_EXT_ADDRESS_DST-1];
	if (addr == NULL)
		return NULL;

	family = ((struct sockaddr *)(addr + 1))->sa_family;
	switch (family) {
	case AF_INET:
		xaddr = (xfrm_address_t *)&((struct sockaddr_in *)(addr + 1))->sin_addr;
		break;
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	case AF_INET6:
		xaddr = (xfrm_address_t *)&((struct sockaddr_in6 *)(addr + 1))->sin6_addr;
		break;
#endif
	default:
		xaddr = NULL;
	}

	if (!xaddr)
		return NULL;

	return xfrm_state_lookup(net, xaddr, sa->sadb_sa_spi, proto, family);
}

#define PFKEY_ALIGN8(a) (1 + (((a) - 1) | (8 - 1)))

static int
pfkey_sockaddr_size(sa_family_t family)
{
	return PFKEY_ALIGN8(pfkey_sockaddr_len(family));
}

static inline int pfkey_mode_from_xfrm(int mode)
{
	switch(mode) {
	case XFRM_MODE_TRANSPORT:
		return IPSEC_MODE_TRANSPORT;
	case XFRM_MODE_TUNNEL:
		return IPSEC_MODE_TUNNEL;
	case XFRM_MODE_BEET:
		return IPSEC_MODE_BEET;
	default:
		return -1;
	}
}

static inline int pfkey_mode_to_xfrm(int mode)
{
	switch(mode) {
	case IPSEC_MODE_ANY:	/*XXX*/
	case IPSEC_MODE_TRANSPORT:
		return XFRM_MODE_TRANSPORT;
	case IPSEC_MODE_TUNNEL:
		return XFRM_MODE_TUNNEL;
	case IPSEC_MODE_BEET:
		return XFRM_MODE_BEET;
	default:
		return -1;
	}
}

static unsigned int pfkey_sockaddr_fill(xfrm_address_t *xaddr, __be16 port,
				       struct sockaddr *sa,
				       unsigned short family)
{
	switch (family) {
	case AF_INET:
	    {
		struct sockaddr_in *sin = (struct sockaddr_in *)sa;
		sin->sin_family = AF_INET;
		sin->sin_port = port;
		sin->sin_addr.s_addr = xaddr->a4;
		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
		return 32;
	    }
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	case AF_INET6:
	    {
		struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sa;
		sin6->sin6_family = AF_INET6;
		sin6->sin6_port = port;
		sin6->sin6_flowinfo = 0;
		ipv6_addr_copy(&sin6->sin6_addr, (struct in6_addr *)xaddr->a6);
		sin6->sin6_scope_id = 0;
		return 128;
	    }
#endif
	}
	return 0;
}

static struct sk_buff *__pfkey_xfrm_state2msg(struct xfrm_state *x,
					      int add_keys, int hsc)
{
	struct sk_buff *skb;
	struct sadb_msg *hdr;
	struct sadb_sa *sa;
	struct sadb_lifetime *lifetime;
	struct sadb_address *addr;
	struct sadb_key *key;
	struct sadb_x_sa2 *sa2;
	struct sadb_x_sec_ctx *sec_ctx;
	struct xfrm_sec_ctx *xfrm_ctx;
	int ctx_size = 0;
	int size;
	int auth_key_size = 0;
	int encrypt_key_size = 0;
	int sockaddr_size;
	struct xfrm_encap_tmpl *natt = NULL;
	int mode;

	/* address family check */
	sockaddr_size = pfkey_sockaddr_size(x->props.family);
	if (!sockaddr_size)
		return ERR_PTR(-EINVAL);

	/* base, SA, (lifetime (HSC),) address(SD), (address(P),)
	   key(AE), (identity(SD),) (sensitivity)> */
	size = sizeof(struct sadb_msg) +sizeof(struct sadb_sa) +
		sizeof(struct sadb_lifetime) +
		((hsc & 1) ? sizeof(struct sadb_lifetime) : 0) +
		((hsc & 2) ? sizeof(struct sadb_lifetime) : 0) +
			sizeof(struct sadb_address)*2 +
				sockaddr_size*2 +
					sizeof(struct sadb_x_sa2);

	if ((xfrm_ctx = x->security)) {
		ctx_size = PFKEY_ALIGN8(xfrm_ctx->ctx_len);
		size += sizeof(struct sadb_x_sec_ctx) + ctx_size;
	}

	/* identity & sensitivity */
	if (xfrm_addr_cmp(&x->sel.saddr, &x->props.saddr, x->props.family))
		size += sizeof(struct sadb_address) + sockaddr_size;

	if (add_keys) {
		if (x->aalg && x->aalg->alg_key_len) {
			auth_key_size =
				PFKEY_ALIGN8((x->aalg->alg_key_len + 7) / 8);
			size += sizeof(struct sadb_key) + auth_key_size;
		}
		if (x->ealg && x->ealg->alg_key_len) {
			encrypt_key_size =
				PFKEY_ALIGN8((x->ealg->alg_key_len+7) / 8);
			size += sizeof(struct sadb_key) + encrypt_key_size;
		}
	}
	if (x->encap)
		natt = x->encap;

	if (natt && natt->encap_type) {
		size += sizeof(struct sadb_x_nat_t_type);
		size += sizeof(struct sadb_x_nat_t_port);
		size += sizeof(struct sadb_x_nat_t_port);
	}

	skb =  alloc_skb(size + 16, GFP_ATOMIC);
	if (skb == NULL)
		return ERR_PTR(-ENOBUFS);

	/* call should fill header later */
	hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg));
	memset(hdr, 0, size);	/* XXX do we need this ? */
	hdr->sadb_msg_len = size / sizeof(uint64_t);

	/* sa */
	sa = (struct sadb_sa *)  skb_put(skb, sizeof(struct sadb_sa));
	sa->sadb_sa_len = sizeof(struct sadb_sa)/sizeof(uint64_t);
	sa->sadb_sa_exttype = SADB_EXT_SA;
	sa->sadb_sa_spi = x->id.spi;
	sa->sadb_sa_replay = x->props.replay_window;
	switch (x->km.state) {
	case XFRM_STATE_VALID:
		sa->sadb_sa_state = x->km.dying ?
			SADB_SASTATE_DYING : SADB_SASTATE_MATURE;
		break;
	case XFRM_STATE_ACQ:
		sa->sadb_sa_state = SADB_SASTATE_LARVAL;
		break;
	default:
		sa->sadb_sa_state = SADB_SASTATE_DEAD;
		break;
	}
	sa->sadb_sa_auth = 0;
	if (x->aalg) {
		struct xfrm_algo_desc *a = xfrm_aalg_get_byname(x->aalg->alg_name, 0);
		sa->sadb_sa_auth = a ? a->desc.sadb_alg_id : 0;
	}
	sa->sadb_sa_encrypt = 0;
	BUG_ON(x->ealg && x->calg);
	if (x->ealg) {
		struct xfrm_algo_desc *a = xfrm_ealg_get_byname(x->ealg->alg_name, 0);
		sa->sadb_sa_encrypt = a ? a->desc.sadb_alg_id : 0;
	}
	/* KAME compatible: sadb_sa_encrypt is overloaded with calg id */
	if (x->calg) {
		struct xfrm_algo_desc *a = xfrm_calg_get_byname(x->calg->alg_name, 0);
		sa->sadb_sa_encrypt = a ? a->desc.sadb_alg_id : 0;
	}

	sa->sadb_sa_flags = 0;
	if (x->props.flags & XFRM_STATE_NOECN)
		sa->sadb_sa_flags |= SADB_SAFLAGS_NOECN;
	if (x->props.flags & XFRM_STATE_DECAP_DSCP)
		sa->sadb_sa_flags |= SADB_SAFLAGS_DECAP_DSCP;
	if (x->props.flags & XFRM_STATE_NOPMTUDISC)
		sa->sadb_sa_flags |= SADB_SAFLAGS_NOPMTUDISC;

	/* hard time */
	if (hsc & 2) {
		lifetime = (struct sadb_lifetime *)  skb_put(skb,
							     sizeof(struct sadb_lifetime));
		lifetime->sadb_lifetime_len =
			sizeof(struct sadb_lifetime)/sizeof(uint64_t);
		lifetime->sadb_lifetime_exttype = SADB_EXT_LIFETIME_HARD;
		lifetime->sadb_lifetime_allocations =  _X2KEY(x->lft.hard_packet_limit);
		lifetime->sadb_lifetime_bytes = _X2KEY(x->lft.hard_byte_limit);
		lifetime->sadb_lifetime_addtime = x->lft.hard_add_expires_seconds;
		lifetime->sadb_lifetime_usetime = x->lft.hard_use_expires_seconds;
	}
	/* soft time */
	if (hsc & 1) {
		lifetime = (struct sadb_lifetime *)  skb_put(skb,
							     sizeof(struct sadb_lifetime));
		lifetime->sadb_lifetime_len =
			sizeof(struct sadb_lifetime)/sizeof(uint64_t);
		lifetime->sadb_lifetime_exttype = SADB_EXT_LIFETIME_SOFT;
		lifetime->sadb_lifetime_allocations =  _X2KEY(x->lft.soft_packet_limit);
		lifetime->sadb_lifetime_bytes = _X2KEY(x->lft.soft_byte_limit);
		lifetime->sadb_lifetime_addtime = x->lft.soft_add_expires_seconds;
		lifetime->sadb_lifetime_usetime = x->lft.soft_use_expires_seconds;
	}
	/* current time */
	lifetime = (struct sadb_lifetime *)  skb_put(skb,
						     sizeof(struct sadb_lifetime));
	lifetime->sadb_lifetime_len =
		sizeof(struct sadb_lifetime)/sizeof(uint64_t);
	lifetime->sadb_lifetime_exttype = SADB_EXT_LIFETIME_CURRENT;
	lifetime->sadb_lifetime_allocations = x->curlft.packets;
	lifetime->sadb_lifetime_bytes = x->curlft.bytes;
	lifetime->sadb_lifetime_addtime = x->curlft.add_time;
	lifetime->sadb_lifetime_usetime = x->curlft.use_time;
	/* src address */
	addr = (struct sadb_address*) skb_put(skb,
					      sizeof(struct sadb_address)+sockaddr_size);
	addr->sadb_address_len =
		(sizeof(struct sadb_address)+sockaddr_size)/
			sizeof(uint64_t);
	addr->sadb_address_exttype = SADB_EXT_ADDRESS_SRC;
	/* "if the ports are non-zero, then the sadb_address_proto field,
	   normally zero, MUST be filled in with the transport
	   protocol's number." - RFC2367 */
	addr->sadb_address_proto = 0;
	addr->sadb_address_reserved = 0;

	addr->sadb_address_prefixlen =
		pfkey_sockaddr_fill(&x->props.saddr, 0,
				    (struct sockaddr *) (addr + 1),
				    x->props.family);
	if (!addr->sadb_address_prefixlen)
		BUG();

	/* dst address */
	addr = (struct sadb_address*) skb_put(skb,
					      sizeof(struct sadb_address)+sockaddr_size);
	addr->sadb_address_len =
		(sizeof(struct sadb_address)+sockaddr_size)/
			sizeof(uint64_t);
	addr->sadb_address_exttype = SADB_EXT_ADDRESS_DST;
	addr->sadb_address_proto = 0;
	addr->sadb_address_reserved = 0;

	addr->sadb_address_prefixlen =
		pfkey_sockaddr_fill(&x->id.daddr, 0,
				    (struct sockaddr *) (addr + 1),
				    x->props.family);
	if (!addr->sadb_address_prefixlen)
		BUG();

	if (xfrm_addr_cmp(&x->sel.saddr, &x->props.saddr,
			  x->props.family)) {
		addr = (struct sadb_address*) skb_put(skb,
			sizeof(struct sadb_address)+sockaddr_size);
		addr->sadb_address_len =
			(sizeof(struct sadb_address)+sockaddr_size)/
			sizeof(uint64_t);
		addr->sadb_address_exttype = SADB_EXT_ADDRESS_PROXY;
		addr->sadb_address_proto =
			pfkey_proto_from_xfrm(x->sel.proto);
		addr->sadb_address_prefixlen = x->sel.prefixlen_s;
		addr->sadb_address_reserved = 0;

		pfkey_sockaddr_fill(&x->sel.saddr, x->sel.sport,
				    (struct sockaddr *) (addr + 1),
				    x->props.family);
	}

	/* auth key */
	if (add_keys && auth_key_size) {
		key = (struct sadb_key *) skb_put(skb,
						  sizeof(struct sadb_key)+auth_key_size);
		key->sadb_key_len = (sizeof(struct sadb_key) + auth_key_size) /
			sizeof(uint64_t);
		key->sadb_key_exttype = SADB_EXT_KEY_AUTH;
		key->sadb_key_bits = x->aalg->alg_key_len;
		key->sadb_key_reserved = 0;
		memcpy(key + 1, x->aalg->alg_key, (x->aalg->alg_key_len+7)/8);
	}
	/* encrypt key */
	if (add_keys && encrypt_key_size) {
		key = (struct sadb_key *) skb_put(skb,
						  sizeof(struct sadb_key)+encrypt_key_size);
		key->sadb_key_len = (sizeof(struct sadb_key) +
				     encrypt_key_size) / sizeof(uint64_t);
		key->sadb_key_exttype = SADB_EXT_KEY_ENCRYPT;
		key->sadb_key_bits = x->ealg->alg_key_len;
		key->sadb_key_reserved = 0;
		memcpy(key + 1, x->ealg->alg_key,
		       (x->ealg->alg_key_len+7)/8);
	}

	/* sa */
	sa2 = (struct sadb_x_sa2 *)  skb_put(skb, sizeof(struct sadb_x_sa2));
	sa2->sadb_x_sa2_len = sizeof(struct sadb_x_sa2)/sizeof(uint64_t);
	sa2->sadb_x_sa2_exttype = SADB_X_EXT_SA2;
	if ((mode = pfkey_mode_from_xfrm(x->props.mode)) < 0) {
		kfree_skb(skb);
		return ERR_PTR(-EINVAL);
	}
	sa2->sadb_x_sa2_mode = mode;
	sa2->sadb_x_sa2_reserved1 = 0;
	sa2->sadb_x_sa2_reserved2 = 0;
	sa2->sadb_x_sa2_sequence = 0;
	sa2->sadb_x_sa2_reqid = x->props.reqid;

	if (natt && natt->encap_type) {
		struct sadb_x_nat_t_type *n_type;
		struct sadb_x_nat_t_port *n_port;

		/* type */
		n_type = (struct sadb_x_nat_t_type*) skb_put(skb, sizeof(*n_type));
		n_type->sadb_x_nat_t_type_len = sizeof(*n_type)/sizeof(uint64_t);
		n_type->sadb_x_nat_t_type_exttype = SADB_X_EXT_NAT_T_TYPE;
		n_type->sadb_x_nat_t_type_type = natt->encap_type;
		n_type->sadb_x_nat_t_type_reserved[0] = 0;
		n_type->sadb_x_nat_t_type_reserved[1] = 0;
		n_type->sadb_x_nat_t_type_reserved[2] = 0;

		/* source port */
		n_port = (struct sadb_x_nat_t_port*) skb_put(skb, sizeof (*n_port));
		n_port->sadb_x_nat_t_port_len = sizeof(*n_port)/sizeof(uint64_t);
		n_port->sadb_x_nat_t_port_exttype = SADB_X_EXT_NAT_T_SPORT;
		n_port->sadb_x_nat_t_port_port = natt->encap_sport;
		n_port->sadb_x_nat_t_port_reserved = 0;

		/* dest port */
		n_port = (struct sadb_x_nat_t_port*) skb_put(skb, sizeof (*n_port));
		n_port->sadb_x_nat_t_port_len = sizeof(*n_port)/sizeof(uint64_t);
		n_port->sadb_x_nat_t_port_exttype = SADB_X_EXT_NAT_T_DPORT;
		n_port->sadb_x_nat_t_port_port = natt->encap_dport;
		n_port->sadb_x_nat_t_port_reserved = 0;
	}

	/* security context */
	if (xfrm_ctx) {
		sec_ctx = (struct sadb_x_sec_ctx *) skb_put(skb,
				sizeof(struct sadb_x_sec_ctx) + ctx_size);
		sec_ctx->sadb_x_sec_len =
		  (sizeof(struct sadb_x_sec_ctx) + ctx_size) / sizeof(uint64_t);
		sec_ctx->sadb_x_sec_exttype = SADB_X_EXT_SEC_CTX;
		sec_ctx->sadb_x_ctx_doi = xfrm_ctx->ctx_doi;
		sec_ctx->sadb_x_ctx_alg = xfrm_ctx->ctx_alg;
		sec_ctx->sadb_x_ctx_len = xfrm_ctx->ctx_len;
		memcpy(sec_ctx + 1, xfrm_ctx->ctx_str,
		       xfrm_ctx->ctx_len);
	}

	return skb;
}


static inline struct sk_buff *pfkey_xfrm_state2msg(struct xfrm_state *x)
{
	struct sk_buff *skb;

	skb = __pfkey_xfrm_state2msg(x, 1, 3);

	return skb;
}

static inline struct sk_buff *pfkey_xfrm_state2msg_expire(struct xfrm_state *x,
							  int hsc)
{
	return __pfkey_xfrm_state2msg(x, 0, hsc);
}

static struct xfrm_state * pfkey_msg2xfrm_state(struct net *net,
						struct sadb_msg *hdr,
						void **ext_hdrs)
{
	struct xfrm_state *x;
	struct sadb_lifetime *lifetime;
	struct sadb_sa *sa;
	struct sadb_key *key;
	struct sadb_x_sec_ctx *sec_ctx;
	uint16_t proto;
	int err;


	sa = (struct sadb_sa *) ext_hdrs[SADB_EXT_SA-1];
	if (!sa ||
	    !present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1],
				     ext_hdrs[SADB_EXT_ADDRESS_DST-1]))
		return ERR_PTR(-EINVAL);
	if (hdr->sadb_msg_satype == SADB_SATYPE_ESP &&
	    !ext_hdrs[SADB_EXT_KEY_ENCRYPT-1])
		return ERR_PTR(-EINVAL);
	if (hdr->sadb_msg_satype == SADB_SATYPE_AH &&
	    !ext_hdrs[SADB_EXT_KEY_AUTH-1])
		return ERR_PTR(-EINVAL);
	if (!!ext_hdrs[SADB_EXT_LIFETIME_HARD-1] !=
	    !!ext_hdrs[SADB_EXT_LIFETIME_SOFT-1])
		return ERR_PTR(-EINVAL);

	proto = pfkey_satype2proto(hdr->sadb_msg_satype);
	if (proto == 0)
		return ERR_PTR(-EINVAL);

	/* default error is no buffer space */
	err = -ENOBUFS;

	/* RFC2367:

   Only SADB_SASTATE_MATURE SAs may be submitted in an SADB_ADD message.
   SADB_SASTATE_LARVAL SAs are created by SADB_GETSPI and it is not
   sensible to add a new SA in the DYING or SADB_SASTATE_DEAD state.
   Therefore, the sadb_sa_state field of all submitted SAs MUST be
   SADB_SASTATE_MATURE and the kernel MUST return an error if this is
   not true.

	   However, KAME setkey always uses SADB_SASTATE_LARVAL.
	   Hence, we have to _ignore_ sadb_sa_state, which is also reasonable.
	 */
	if (sa->sadb_sa_auth > SADB_AALG_MAX ||
	    (hdr->sadb_msg_satype == SADB_X_SATYPE_IPCOMP &&
	     sa->sadb_sa_encrypt > SADB_X_CALG_MAX) ||
	    sa->sadb_sa_encrypt > SADB_EALG_MAX)
		return ERR_PTR(-EINVAL);
	key = (struct sadb_key*) ext_hdrs[SADB_EXT_KEY_AUTH-1];
	if (key != NULL &&
	    sa->sadb_sa_auth != SADB_X_AALG_NULL &&
	    ((key->sadb_key_bits+7) / 8 == 0 ||
	     (key->sadb_key_bits+7) / 8 > key->sadb_key_len * sizeof(uint64_t)))
		return ERR_PTR(-EINVAL);
	key = ext_hdrs[SADB_EXT_KEY_ENCRYPT-1];
	if (key != NULL &&
	    sa->sadb_sa_encrypt != SADB_EALG_NULL &&
	    ((key->sadb_key_bits+7) / 8 == 0 ||
	     (key->sadb_key_bits+7) / 8 > key->sadb_key_len * sizeof(uint64_t)))
		return ERR_PTR(-EINVAL);

	x = xfrm_state_alloc(net);
	if (x == NULL)
		return ERR_PTR(-ENOBUFS);

	x->id.proto = proto;
	x->id.spi = sa->sadb_sa_spi;
	x->props.replay_window = sa->sadb_sa_replay;
	if (sa->sadb_sa_flags & SADB_SAFLAGS_NOECN)
		x->props.flags |= XFRM_STATE_NOECN;
	if (sa->sadb_sa_flags & SADB_SAFLAGS_DECAP_DSCP)
		x->props.flags |= XFRM_STATE_DECAP_DSCP;
	if (sa->sadb_sa_flags & SADB_SAFLAGS_NOPMTUDISC)
		x->props.flags |= XFRM_STATE_NOPMTUDISC;

	lifetime = (struct sadb_lifetime*) ext_hdrs[SADB_EXT_LIFETIME_HARD-1];
	if (lifetime != NULL) {
		x->lft.hard_packet_limit = _KEY2X(lifetime->sadb_lifetime_allocations);
		x->lft.hard_byte_limit = _KEY2X(lifetime->sadb_lifetime_bytes);
		x->lft.hard_add_expires_seconds = lifetime->sadb_lifetime_addtime;
		x->lft.hard_use_expires_seconds = lifetime->sadb_lifetime_usetime;
	}
	lifetime = (struct sadb_lifetime*) ext_hdrs[SADB_EXT_LIFETIME_SOFT-1];
	if (lifetime != NULL) {
		x->lft.soft_packet_limit = _KEY2X(lifetime->sadb_lifetime_allocations);
		x->lft.soft_byte_limit = _KEY2X(lifetime->sadb_lifetime_bytes);
		x->lft.soft_add_expires_seconds = lifetime->sadb_lifetime_addtime;
		x->lft.soft_use_expires_seconds = lifetime->sadb_lifetime_usetime;
	}

	sec_ctx = (struct sadb_x_sec_ctx *) ext_hdrs[SADB_X_EXT_SEC_CTX-1];
	if (sec_ctx != NULL) {
		struct xfrm_user_sec_ctx *uctx = pfkey_sadb2xfrm_user_sec_ctx(sec_ctx);

		if (!uctx)
			goto out;

		err = security_xfrm_state_alloc(x, uctx);
		kfree(uctx);

		if (err)
			goto out;
	}

	key = (struct sadb_key*) ext_hdrs[SADB_EXT_KEY_AUTH-1];
	if (sa->sadb_sa_auth) {
		int keysize = 0;
		struct xfrm_algo_desc *a = xfrm_aalg_get_byid(sa->sadb_sa_auth);
		if (!a) {
			err = -ENOSYS;
			goto out;
		}
		if (key)
			keysize = (key->sadb_key_bits + 7) / 8;
		x->aalg = kmalloc(sizeof(*x->aalg) + keysize, GFP_KERNEL);
		if (!x->aalg)
			goto out;
		strcpy(x->aalg->alg_name, a->name);
		x->aalg->alg_key_len = 0;
		if (key) {
			x->aalg->alg_key_len = key->sadb_key_bits;
			memcpy(x->aalg->alg_key, key+1, keysize);
		}
		x->props.aalgo = sa->sadb_sa_auth;
		/* x->algo.flags = sa->sadb_sa_flags; */
	}
	if (sa->sadb_sa_encrypt) {
		if (hdr->sadb_msg_satype == SADB_X_SATYPE_IPCOMP) {
			struct xfrm_algo_desc *a = xfrm_calg_get_byid(sa->sadb_sa_encrypt);
			if (!a) {
				err = -ENOSYS;
				goto out;
			}
			x->calg = kmalloc(sizeof(*x->calg), GFP_KERNEL);
			if (!x->calg)
				goto out;
			strcpy(x->calg->alg_name, a->name);
			x->props.calgo = sa->sadb_sa_encrypt;
		} else {
			int keysize = 0;
			struct xfrm_algo_desc *a = xfrm_ealg_get_byid(sa->sadb_sa_encrypt);
			if (!a) {
				err = -ENOSYS;
				goto out;
			}
			key = (struct sadb_key*) ext_hdrs[SADB_EXT_KEY_ENCRYPT-1];
			if (key)
				keysize = (key->sadb_key_bits + 7) / 8;
			x->ealg = kmalloc(sizeof(*x->ealg) + keysize, GFP_KERNEL);
			if (!x->ealg)
				goto out;
			strcpy(x->ealg->alg_name, a->name);
			x->ealg->alg_key_len = 0;
			if (key) {
				x->ealg->alg_key_len = key->sadb_key_bits;
				memcpy(x->ealg->alg_key, key+1, keysize);
			}
			x->props.ealgo = sa->sadb_sa_encrypt;
		}
	}
	/* x->algo.flags = sa->sadb_sa_flags; */

	x->props.family = pfkey_sadb_addr2xfrm_addr((struct sadb_address *) ext_hdrs[SADB_EXT_ADDRESS_SRC-1],
						    &x->props.saddr);
	if (!x->props.family) {
		err = -EAFNOSUPPORT;
		goto out;
	}
	pfkey_sadb_addr2xfrm_addr((struct sadb_address *) ext_hdrs[SADB_EXT_ADDRESS_DST-1],
				  &x->id.daddr);

	if (ext_hdrs[SADB_X_EXT_SA2-1]) {
		struct sadb_x_sa2 *sa2 = (void*)ext_hdrs[SADB_X_EXT_SA2-1];
		int mode = pfkey_mode_to_xfrm(sa2->sadb_x_sa2_mode);
		if (mode < 0) {
			err = -EINVAL;
			goto out;
		}
		x->props.mode = mode;
		x->props.reqid = sa2->sadb_x_sa2_reqid;
	}

	if (ext_hdrs[SADB_EXT_ADDRESS_PROXY-1]) {
		struct sadb_address *addr = ext_hdrs[SADB_EXT_ADDRESS_PROXY-1];

		/* Nobody uses this, but we try. */
		x->sel.family = pfkey_sadb_addr2xfrm_addr(addr, &x->sel.saddr);
		x->sel.prefixlen_s = addr->sadb_address_prefixlen;
	}

	if (!x->sel.family)
		x->sel.family = x->props.family;

	if (ext_hdrs[SADB_X_EXT_NAT_T_TYPE-1]) {
		struct sadb_x_nat_t_type* n_type;
		struct xfrm_encap_tmpl *natt;

		x->encap = kmalloc(sizeof(*x->encap), GFP_KERNEL);
		if (!x->encap)
			goto out;

		natt = x->encap;
		n_type = ext_hdrs[SADB_X_EXT_NAT_T_TYPE-1];
		natt->encap_type = n_type->sadb_x_nat_t_type_type;

		if (ext_hdrs[SADB_X_EXT_NAT_T_SPORT-1]) {
			struct sadb_x_nat_t_port* n_port =
				ext_hdrs[SADB_X_EXT_NAT_T_SPORT-1];
			natt->encap_sport = n_port->sadb_x_nat_t_port_port;
		}
		if (ext_hdrs[SADB_X_EXT_NAT_T_DPORT-1]) {
			struct sadb_x_nat_t_port* n_port =
				ext_hdrs[SADB_X_EXT_NAT_T_DPORT-1];
			natt->encap_dport = n_port->sadb_x_nat_t_port_port;
		}
		memset(&natt->encap_oa, 0, sizeof(natt->encap_oa));
	}

	err = xfrm_init_state(x);
	if (err)
		goto out;

	x->km.seq = hdr->sadb_msg_seq;
	return x;

out:
	x->km.state = XFRM_STATE_DEAD;
	xfrm_state_put(x);
	return ERR_PTR(err);
}

static int pfkey_reserved(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	return -EOPNOTSUPP;
}

static int pfkey_getspi(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct net *net = sock_net(sk);
	struct sk_buff *resp_skb;
	struct sadb_x_sa2 *sa2;
	struct sadb_address *saddr, *daddr;
	struct sadb_msg *out_hdr;
	struct sadb_spirange *range;
	struct xfrm_state *x = NULL;
	int mode;
	int err;
	u32 min_spi, max_spi;
	u32 reqid;
	u8 proto;
	unsigned short family;
	xfrm_address_t *xsaddr = NULL, *xdaddr = NULL;

	if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1],
				     ext_hdrs[SADB_EXT_ADDRESS_DST-1]))
		return -EINVAL;

	proto = pfkey_satype2proto(hdr->sadb_msg_satype);
	if (proto == 0)
		return -EINVAL;

	if ((sa2 = ext_hdrs[SADB_X_EXT_SA2-1]) != NULL) {
		mode = pfkey_mode_to_xfrm(sa2->sadb_x_sa2_mode);
		if (mode < 0)
			return -EINVAL;
		reqid = sa2->sadb_x_sa2_reqid;
	} else {
		mode = 0;
		reqid = 0;
	}

	saddr = ext_hdrs[SADB_EXT_ADDRESS_SRC-1];
	daddr = ext_hdrs[SADB_EXT_ADDRESS_DST-1];

	family = ((struct sockaddr *)(saddr + 1))->sa_family;
	switch (family) {
	case AF_INET:
		xdaddr = (xfrm_address_t *)&((struct sockaddr_in *)(daddr + 1))->sin_addr.s_addr;
		xsaddr = (xfrm_address_t *)&((struct sockaddr_in *)(saddr + 1))->sin_addr.s_addr;
		break;
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
	case AF_INET6:
		xdaddr = (xfrm_address_t *)&((struct sockaddr_in6 *)(daddr + 1))->sin6_addr;
		xsaddr = (xfrm_address_t *)&((struct sockaddr_in6 *)(saddr + 1))->sin6_addr;
		break;
#endif
	}

	if (hdr->sadb_msg_seq) {
		x = xfrm_find_acq_byseq(net, hdr->sadb_msg_seq);
		if (x && xfrm_addr_cmp(&x->id.daddr, xdaddr, family)) {
			xfrm_state_put(x);
			x = NULL;
		}
	}

	if (!x)
		x = xfrm_find_acq(net, mode, reqid, proto, xdaddr, xsaddr, 1, family);

	if (x == NULL)
		return -ENOENT;

	min_spi = 0x100;
	max_spi = 0x0fffffff;

	range = ext_hdrs[SADB_EXT_SPIRANGE-1];
	if (range) {
		min_spi = range->sadb_spirange_min;
		max_spi = range->sadb_spirange_max;
	}

	err = xfrm_alloc_spi(x, min_spi, max_spi);
	resp_skb = err ? ERR_PTR(err) : pfkey_xfrm_state2msg(x);

	if (IS_ERR(resp_skb)) {
		xfrm_state_put(x);
		return  PTR_ERR(resp_skb);
	}

	out_hdr = (struct sadb_msg *) resp_skb->data;
	out_hdr->sadb_msg_version = hdr->sadb_msg_version;
	out_hdr->sadb_msg_type = SADB_GETSPI;
	out_hdr->sadb_msg_satype = pfkey_proto2satype(proto);
	out_hdr->sadb_msg_errno = 0;
	out_hdr->sadb_msg_reserved = 0;
	out_hdr->sadb_msg_seq = hdr->sadb_msg_seq;
	out_hdr->sadb_msg_pid = hdr->sadb_msg_pid;

	xfrm_state_put(x);

	pfkey_broadcast(resp_skb, GFP_KERNEL, BROADCAST_ONE, sk, net);

	return 0;
}

static int pfkey_acquire(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct net *net = sock_net(sk);
	struct xfrm_state *x;

	if (hdr->sadb_msg_len != sizeof(struct sadb_msg)/8)
		return -EOPNOTSUPP;

	if (hdr->sadb_msg_seq == 0 || hdr->sadb_msg_errno == 0)
		return 0;

	x = xfrm_find_acq_byseq(net, hdr->sadb_msg_seq);
	if (x == NULL)
		return 0;

	spin_lock_bh(&x->lock);
	if (x->km.state == XFRM_STATE_ACQ) {
		x->km.state = XFRM_STATE_ERROR;
		wake_up(&net->xfrm.km_waitq);
	}
	spin_unlock_bh(&x->lock);
	xfrm_state_put(x);
	return 0;
}

static inline int event2poltype(int event)
{
	switch (event) {
	case XFRM_MSG_DELPOLICY:
		return SADB_X_SPDDELETE;
	case XFRM_MSG_NEWPOLICY:
		return SADB_X_SPDADD;
	case XFRM_MSG_UPDPOLICY:
		return SADB_X_SPDUPDATE;
	case XFRM_MSG_POLEXPIRE:
	//	return SADB_X_SPDEXPIRE;
	default:
		printk("pfkey: Unknown policy event %d\n", event);
		break;
	}

	return 0;
}

static inline int event2keytype(int event)
{
	switch (event) {
	case XFRM_MSG_DELSA:
		return SADB_DELETE;
	case XFRM_MSG_NEWSA:
		return SADB_ADD;
	case XFRM_MSG_UPDSA:
		return SADB_UPDATE;
	case XFRM_MSG_EXPIRE:
		return SADB_EXPIRE;
	default:
		printk("pfkey: Unknown SA event %d\n", event);
		break;
	}

	return 0;
}

/* ADD/UPD/DEL */
static int key_notify_sa(struct xfrm_state *x, struct km_event *c)
{
	struct sk_buff *skb;
	struct sadb_msg *hdr;

	skb = pfkey_xfrm_state2msg(x);

	if (IS_ERR(skb))
		return PTR_ERR(skb);

	hdr = (struct sadb_msg *) skb->data;
	hdr->sadb_msg_version = PF_KEY_V2;
	hdr->sadb_msg_type = event2keytype(c->event);
	hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto);
	hdr->sadb_msg_errno = 0;
	hdr->sadb_msg_reserved = 0;
	hdr->sadb_msg_seq = c->seq;
	hdr->sadb_msg_pid = c->pid;

	pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, xs_net(x));

	return 0;
}

static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct net *net = sock_net(sk);
	struct xfrm_state *x;
	int err;
	struct km_event c;

	x = pfkey_msg2xfrm_state(net, hdr, ext_hdrs);
	if (IS_ERR(x))
		return PTR_ERR(x);

	xfrm_state_hold(x);
	if (hdr->sadb_msg_type == SADB_ADD)
		err = xfrm_state_add(x);
	else
		err = xfrm_state_update(x);

	xfrm_audit_state_add(x, err ? 0 : 1,
			     audit_get_loginuid(current),
			     audit_get_sessionid(current), 0);

	if (err < 0) {
		x->km.state = XFRM_STATE_DEAD;
		__xfrm_state_put(x);
		goto out;
	}

	if (hdr->sadb_msg_type == SADB_ADD)
		c.event = XFRM_MSG_NEWSA;
	else
		c.event = XFRM_MSG_UPDSA;
	c.seq = hdr->sadb_msg_seq;
	c.pid = hdr->sadb_msg_pid;
	km_state_notify(x, &c);
out:
	xfrm_state_put(x);
	return err;
}

static int pfkey_delete(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct net *net = sock_net(sk);
	struct xfrm_state *x;
	struct km_event c;
	int err;

	if (!ext_hdrs[SADB_EXT_SA-1] ||
	    !present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1],
				     ext_hdrs[SADB_EXT_ADDRESS_DST-1]))
		return -EINVAL;

	x = pfkey_xfrm_state_lookup(net, hdr, ext_hdrs);
	if (x == NULL)
		return -ESRCH;

	if ((err = security_xfrm_state_delete(x)))
		goto out;

	if (xfrm_state_kern(x)) {
		err = -EPERM;
		goto out;
	}

	err = xfrm_state_delete(x);

	if (err < 0)
		goto out;

	c.seq = hdr->sadb_msg_seq;
	c.pid = hdr->sadb_msg_pid;
	c.event = XFRM_MSG_DELSA;
	km_state_notify(x, &c);
out:
	xfrm_audit_state_delete(x, err ? 0 : 1,
				audit_get_loginuid(current),
				audit_get_sessionid(current), 0);
	xfrm_state_put(x);

	return err;
}

static int pfkey_get(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct net *net = sock_net(sk);
	__u8 proto;
	struct sk_buff *out_skb;
	struct sadb_msg *out_hdr;
	struct xfrm_state *x;

	if (!ext_hdrs[SADB_EXT_SA-1] ||
	    !present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1],
				     ext_hdrs[SADB_EXT_ADDRESS_DST-1]))
		return -EINVAL;

	x = pfkey_xfrm_state_lookup(net, hdr, ext_hdrs);
	if (x == NULL)
		return -ESRCH;

	out_skb = pfkey_xfrm_state2msg(x);
	proto = x->id.proto;
	xfrm_state_put(x);
	if (IS_ERR(out_skb))
		return  PTR_ERR(out_skb);

	out_hdr = (struct sadb_msg *) out_skb->data;
	out_hdr->sadb_msg_version = hdr->sadb_msg_version;
	out_hdr->sadb_msg_type = SADB_GET;
	out_hdr->sadb_msg_satype = pfkey_proto2satype(proto);
	out_hdr->sadb_msg_errno = 0;
	out_hdr->sadb_msg_reserved = 0;
	out_hdr->sadb_msg_seq = hdr->sadb_msg_seq;
	out_hdr->sadb_msg_pid = hdr->sadb_msg_pid;
	pfkey_broadcast(out_skb, GFP_ATOMIC, BROADCAST_ONE, sk, sock_net(sk));

	return 0;
}

static struct sk_buff *compose_sadb_supported(struct sadb_msg *orig,
					      gfp_t allocation)
{
	struct sk_buff *skb;
	struct sadb_msg *hdr;
	int len, auth_len, enc_len, i;

	auth_len = xfrm_count_auth_supported();
	if (auth_len) {
		auth_len *= sizeof(struct sadb_alg);
		auth_len += sizeof(struct sadb_supported);
	}

	enc_len = xfrm_count_enc_supported();
	if (enc_len) {
		enc_len *= sizeof(struct sadb_alg);
		enc_len += sizeof(struct sadb_supported);
	}

	len = enc_len + auth_len + sizeof(struct sadb_msg);

	skb = alloc_skb(len + 16, allocation);
	if (!skb)
		goto out_put_algs;

	hdr = (struct sadb_msg *) skb_put(skb, sizeof(*hdr));
	pfkey_hdr_dup(hdr, orig);
	hdr->sadb_msg_errno = 0;
	hdr->sadb_msg_len = len / sizeof(uint64_t);

	if (auth_len) {
		struct sadb_supported *sp;
		struct sadb_alg *ap;

		sp = (struct sadb_supported *) skb_put(skb, auth_len);
		ap = (struct sadb_alg *) (sp + 1);

		sp->sadb_supported_len = auth_len / sizeof(uint64_t);
		sp->sadb_supported_exttype = SADB_EXT_SUPPORTED_AUTH;

		for (i = 0; ; i++) {
			struct xfrm_algo_desc *aalg = xfrm_aalg_get_byidx(i);
			if (!aalg)
				break;
			if (aalg->available)
				*ap++ = aalg->desc;
		}
	}

	if (enc_len) {
		struct sadb_supported *sp;
		struct sadb_alg *ap;

		sp = (struct sadb_supported *) skb_put(skb, enc_len);
		ap = (struct sadb_alg *) (sp + 1);

		sp->sadb_supported_len = enc_len / sizeof(uint64_t);
		sp->sadb_supported_exttype = SADB_EXT_SUPPORTED_ENCRYPT;

		for (i = 0; ; i++) {
			struct xfrm_algo_desc *ealg = xfrm_ealg_get_byidx(i);
			if (!ealg)
				break;
			if (ealg->available)
				*ap++ = ealg->desc;
		}
	}

out_put_algs:
	return skb;
}

static int pfkey_register(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct pfkey_sock *pfk = pfkey_sk(sk);
	struct sk_buff *supp_skb;

	if (hdr->sadb_msg_satype > SADB_SATYPE_MAX)
		return -EINVAL;

	if (hdr->sadb_msg_satype != SADB_SATYPE_UNSPEC) {
		if (pfk->registered&(1<<hdr->sadb_msg_satype))
			return -EEXIST;
		pfk->registered |= (1<<hdr->sadb_msg_satype);
	}

	xfrm_probe_algs();

	supp_skb = compose_sadb_supported(hdr, GFP_KERNEL);
	if (!supp_skb) {
		if (hdr->sadb_msg_satype != SADB_SATYPE_UNSPEC)
			pfk->registered &= ~(1<<hdr->sadb_msg_satype);

		return -ENOBUFS;
	}

	pfkey_broadcast(supp_skb, GFP_KERNEL, BROADCAST_REGISTERED, sk, sock_net(sk));

	return 0;
}

static int key_notify_sa_flush(struct km_event *c)
{
	struct sk_buff *skb;
	struct sadb_msg *hdr;

	skb = alloc_skb(sizeof(struct sadb_msg) + 16, GFP_ATOMIC);
	if (!skb)
		return -ENOBUFS;
	hdr = (struct sadb_msg *) skb_put(skb, sizeof(struct sadb_msg));
	hdr->sadb_msg_satype = pfkey_proto2satype(c->data.proto);
	hdr->sadb_msg_type = SADB_FLUSH;
	hdr->sadb_msg_seq = c->seq;
	hdr->sadb_msg_pid = c->pid;
	hdr->sadb_msg_version = PF_KEY_V2;
	hdr->sadb_msg_errno = (uint8_t) 0;
	hdr->sadb_msg_len = (sizeof(struct sadb_msg) / sizeof(uint64_t));

	pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, c->net);

	return 0;
}

static int pfkey_flush(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct net *net = sock_net(sk);
	unsigned proto;
	struct km_event c;
	struct xfrm_audit audit_info;
	int err;

	proto = pfkey_satype2proto(hdr->sadb_msg_satype);
	if (proto == 0)
		return -EINVAL;

	audit_info.loginuid = audit_get_loginuid(current);
	audit_info.sessionid = audit_get_sessionid(current);
	audit_info.secid = 0;
	err = xfrm_state_flush(net, proto, &audit_info);
	if (err)
		return err;
	c.data.proto = proto;
	c.seq = hdr->sadb_msg_seq;
	c.pid = hdr->sadb_msg_pid;
	c.event = XFRM_MSG_FLUSHSA;
	c.net = net;
	km_state_notify(NULL, &c);

	return 0;
}

static int dump_sa(struct xfrm_state *x, int count, void *ptr)
{
	struct pfkey_sock *pfk = ptr;
	struct sk_buff *out_skb;
	struct sadb_msg *out_hdr;

	if (!pfkey_can_dump(&pfk->sk))
		return -ENOBUFS;

	out_skb = pfkey_xfrm_state2msg(x);
	if (IS_ERR(out_skb))
		return PTR_ERR(out_skb);

	out_hdr = (struct sadb_msg *) out_skb->data;
	out_hdr->sadb_msg_version = pfk->dump.msg_version;
	out_hdr->sadb_msg_type = SADB_DUMP;
	out_hdr->sadb_msg_satype = pfkey_proto2satype(x->id.proto);
	out_hdr->sadb_msg_errno = 0;
	out_hdr->sadb_msg_reserved = 0;
	out_hdr->sadb_msg_seq = count + 1;
	out_hdr->sadb_msg_pid = pfk->dump.msg_pid;

	if (pfk->dump.skb)
		pfkey_broadcast(pfk->dump.skb, GFP_ATOMIC, BROADCAST_ONE,
				&pfk->sk, sock_net(&pfk->sk));
	pfk->dump.skb = out_skb;

	return 0;
}

static int pfkey_dump_sa(struct pfkey_sock *pfk)
{
	struct net *net = sock_net(&pfk->sk);
	return xfrm_state_walk(net, &pfk->dump.u.state, dump_sa, (void *) pfk);
}

static void pfkey_dump_sa_done(struct pfkey_sock *pfk)
{
	xfrm_state_walk_done(&pfk->dump.u.state);
}

static int pfkey_dump(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	u8 proto;
	struct pfkey_sock *pfk = pfkey_sk(sk);

	if (pfk->dump.dump != NULL)
		return -EBUSY;

	proto = pfkey_satype2proto(hdr->sadb_msg_satype);
	if (proto == 0)
		return -EINVAL;

	pfk->dump.msg_version = hdr->sadb_msg_version;
	pfk->dump.msg_pid = hdr->sadb_msg_pid;
	pfk->dump.dump = pfkey_dump_sa;
	pfk->dump.done = pfkey_dump_sa_done;
	xfrm_state_walk_init(&pfk->dump.u.state, proto);

	return pfkey_do_dump(pfk);
}

static int pfkey_promisc(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs)
{
	struct pfkey_sock *pfk = pfkey_sk(sk);
	int satype = hdr->sadb_msg_satype;

	if (hdr->sadb_msg_len == (sizeof(*hdr) / sizeof(uint64_t))) {
		/* XXX we mangle packet... */
		hdr->sadb_msg_errno = 0;
		if (satype != 0 && satype != 1)
			return -EINVAL;
		pfk->promisc = satype;
	}
	pfkey_broadcast(skb_clone(skb, GFP_KERNEL), GFP_KERNEL, BROADCAST_ALL, NULL, sock_net(sk));
	return 0;
}

static int check_reqid(struct xfrm_policy *xp, int dir, int count, void *ptr)
{
	int i;
	u32 reqid = *(u32*)ptr;

	for (i=0; i<xp->xfrm_nr; i++) {
		if (xp->xfrm_vec[i].reqid == reqid)
			return -EEXIST;
	}
	return 0;
}

static u32 gen_reqid(struct net *net)
{
	struct xfrm_policy_walk walk;
	u32 start;
	int rc;
	static u32 reqid = IPSEC_MANUAL_REQID_MAX;

	start = reqid;
	do {
		++reqid;
		if (reqid == 0)
			reqid = IPSEC_MANUAL_REQID_MAX+1;
		xfrm_policy_walk_init(&walk, XFRM_POLICY_TYPE_MAIN);
		rc = xfrm_policy_walk(net, &walk, check_reqid, (void*)&reqid);
		xfrm_policy_walk_done(&walk);
		if (rc != -EEXIST)
			return reqid;
	} while (reqid != start);
	return 0;
}

static int
parse_ipsecrequest(struct xfrm_policy *xp, struct sadb_x_ipsecrequest *rq)
{
	struct net *net = xp_net(xp);
	struct xfrm_tmpl *t = xp->xfrm_vec + xp->xfrm_nr;
	int mode;

	if (xp->xfrm_nr >= XFRM_MAX_DEPTH)
		return -ELOOP;

	if (rq->sadb_x_ipsecrequest_mode == 0)
		return -EINVAL;

	t->id.proto = rq->sadb_x_ipsecrequest_proto; /* XXX check proto */
	if ((mode = pfkey_mode_to_xfrm(rq->sadb_x_ipsecrequest_mode)) < 0)
		return -EINVAL;
	t->mode = mode;
	if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_USE)
		t->optional = 1;
	else if (rq->sadb_x_ipsecrequest_level == IPSEC_LEVEL_UNIQUE) {
		t->reqid = rq->sadb_x_ipsecrequest_reqid;
		if (t->reqid > IPSEC_MANUAL_REQID_MAX)…