[linux-2.6-block.git] / net / ipv6 / xfrm6_policy.c

/*
 * xfrm6_policy.c: based on xfrm4_policy.c
 *
 * Authors:
 *	Mitsuru KANDA @USAGI
 * 	Kazunori MIYAZAWA @USAGI
 * 	Kunihiro Ishiguro <kunihiro@ipinfusion.com>
 * 		IPv6 support
 * 	YOSHIFUJI Hideaki
 * 		Split up af-specific portion
 *
 */

#include <linux/compiler.h>
#include <linux/netdevice.h>
#include <net/addrconf.h>
#include <net/xfrm.h>
#include <net/ip.h>
#include <net/ipv6.h>
#include <net/ip6_route.h>
#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
#include <net/mip6.h>
#endif

static struct dst_ops xfrm6_dst_ops;
static struct xfrm_policy_afinfo xfrm6_policy_afinfo;

static int xfrm6_dst_lookup(struct xfrm_dst **xdst, struct flowi *fl)
{
	struct dst_entry *dst = ip6_route_output(NULL, fl);
	int err = dst->error;
	if (!err)
		*xdst = (struct xfrm_dst *) dst;
	else
		dst_release(dst);
	return err;
}

static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
{
	struct rt6_info *rt;
	struct flowi fl_tunnel = {
		.nl_u = {
			.ip6_u = {
				.daddr = *(struct in6_addr *)&daddr->a6,
			},
		},
	};

	if (!xfrm6_dst_lookup((struct xfrm_dst **)&rt, &fl_tunnel)) {
		ipv6_get_saddr(&rt->u.dst, (struct in6_addr *)&daddr->a6,
			       (struct in6_addr *)&saddr->a6);
		dst_release(&rt->u.dst);
		return 0;
	}
	return -EHOSTUNREACH;
}

static struct dst_entry *
__xfrm6_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
{
	struct dst_entry *dst;

	/* Still not clear if we should set fl->fl6_{src,dst}... */
	read_lock_bh(&policy->lock);
	for (dst = policy->bundles; dst; dst = dst->next) {
		struct xfrm_dst *xdst = (struct xfrm_dst*)dst;
		struct in6_addr fl_dst_prefix, fl_src_prefix;

		ipv6_addr_prefix(&fl_dst_prefix,
				 &fl->fl6_dst,
				 xdst->u.rt6.rt6i_dst.plen);
		ipv6_addr_prefix(&fl_src_prefix,
				 &fl->fl6_src,
				 xdst->u.rt6.rt6i_src.plen);
		if (ipv6_addr_equal(&xdst->u.rt6.rt6i_dst.addr, &fl_dst_prefix) &&
		    ipv6_addr_equal(&xdst->u.rt6.rt6i_src.addr, &fl_src_prefix) &&
		    xfrm_bundle_ok(policy, xdst, fl, AF_INET6,
				   (xdst->u.rt6.rt6i_dst.plen != 128 ||
				    xdst->u.rt6.rt6i_src.plen != 128))) {
			dst_clone(dst);
			break;
		}
	}
	read_unlock_bh(&policy->lock);
	return dst;
}

static inline struct in6_addr*
__xfrm6_bundle_addr_remote(struct xfrm_state *x, struct in6_addr *addr)
{
	return (x->type->remote_addr) ?
		(struct in6_addr*)x->type->remote_addr(x, (xfrm_address_t *)addr) :
		(struct in6_addr*)&x->id.daddr;
}

static inline struct in6_addr*
__xfrm6_bundle_addr_local(struct xfrm_state *x, struct in6_addr *addr)
{
	return (x->type->local_addr) ?
		(struct in6_addr*)x->type->local_addr(x, (xfrm_address_t *)addr) :
		(struct in6_addr*)&x->props.saddr;
}

static inline void
__xfrm6_bundle_len_inc(int *len, int *nflen, struct xfrm_state *x)
{
	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
		*nflen += x->props.header_len;
	else
		*len += x->props.header_len;
}

static inline void
__xfrm6_bundle_len_dec(int *len, int *nflen, struct xfrm_state *x)
{
	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
		*nflen -= x->props.header_len;
	else
		*len -= x->props.header_len;
}

/* Allocate chain of dst_entry's, attach known xfrm's, calculate
 * all the metrics... Shortly, bundle a bundle.
 */

static int
__xfrm6_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int nx,
		      struct flowi *fl, struct dst_entry **dst_p)
{
	struct dst_entry *dst, *dst_prev;
	struct rt6_info *rt0 = (struct rt6_info*)(*dst_p);
	struct rt6_info *rt  = rt0;
	struct flowi fl_tunnel = {
		.nl_u = {
			.ip6_u = {
				.saddr = fl->fl6_src,
				.daddr = fl->fl6_dst,
			}
		}
	};
	int i;
	int err = 0;
	int header_len = 0;
	int nfheader_len = 0;
	int trailer_len = 0;

	dst = dst_prev = NULL;
	dst_hold(&rt->u.dst);

	for (i = 0; i < nx; i++) {
		struct dst_entry *dst1 = dst_alloc(&xfrm6_dst_ops);
		struct xfrm_dst *xdst;

		if (unlikely(dst1 == NULL)) {
			err = -ENOBUFS;
			dst_release(&rt->u.dst);
			goto error;
		}

		if (!dst)
			dst = dst1;
		else {
			dst_prev->child = dst1;
			dst1->flags |= DST_NOHASH;
			dst_clone(dst1);
		}

		xdst = (struct xfrm_dst *)dst1;
		xdst->route = &rt->u.dst;
		xdst->genid = xfrm[i]->genid;
		if (rt->rt6i_node)
			xdst->route_cookie = rt->rt6i_node->fn_sernum;

		dst1->next = dst_prev;
		dst_prev = dst1;

		__xfrm6_bundle_len_inc(&header_len, &nfheader_len, xfrm[i]);
		trailer_len += xfrm[i]->props.trailer_len;

		if (xfrm[i]->props.mode == XFRM_MODE_TUNNEL ||
		    xfrm[i]->props.mode == XFRM_MODE_ROUTEOPTIMIZATION) {
			unsigned short encap_family = xfrm[i]->props.family;
			switch(encap_family) {
			case AF_INET:
				fl_tunnel.fl4_dst = xfrm[i]->id.daddr.a4;
				fl_tunnel.fl4_src = xfrm[i]->props.saddr.a4;
				break;
			case AF_INET6:
				ipv6_addr_copy(&fl_tunnel.fl6_dst, __xfrm6_bundle_addr_remote(xfrm[i], &fl->fl6_dst));

				ipv6_addr_copy(&fl_tunnel.fl6_src, __xfrm6_bundle_addr_local(xfrm[i], &fl->fl6_src));
				break;
			default:
				BUG_ON(1);
			}

			err = xfrm_dst_lookup((struct xfrm_dst **) &rt,
					      &fl_tunnel, encap_family);
			if (err)
				goto error;
		} else
			dst_hold(&rt->u.dst);
	}

	dst_prev->child = &rt->u.dst;
	dst->path = &rt->u.dst;
	if (rt->rt6i_node)
		((struct xfrm_dst *)dst)->path_cookie = rt->rt6i_node->fn_sernum;

	*dst_p = dst;
	dst = dst_prev;

	dst_prev = *dst_p;
	i = 0;
	for (; dst_prev != &rt->u.dst; dst_prev = dst_prev->child) {
		struct xfrm_dst *x = (struct xfrm_dst*)dst_prev;
		struct xfrm_state_afinfo *afinfo;

		dst_prev->xfrm = xfrm[i++];
		dst_prev->dev = rt->u.dst.dev;
		if (rt->u.dst.dev)
			dev_hold(rt->u.dst.dev);
		dst_prev->obsolete	= -1;
		dst_prev->flags	       |= DST_HOST;
		dst_prev->lastuse	= jiffies;
		dst_prev->header_len	= header_len;
		dst_prev->nfheader_len	= nfheader_len;
		dst_prev->trailer_len	= trailer_len;
		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));

		/* Copy neighbour for reachability confirmation */
		dst_prev->neighbour	= neigh_clone(rt->u.dst.neighbour);
		dst_prev->input		= rt->u.dst.input;
		/* XXX: When IPv4 is implemented as module and can be unloaded,
		 * we should manage reference to xfrm4_output in afinfo->output.
		 * Miyazawa
		 */
		afinfo = xfrm_state_get_afinfo(dst_prev->xfrm->props.family);
		if (!afinfo) {
			dst = *dst_p;
			goto error;
		}

		dst_prev->output = afinfo->output;
		xfrm_state_put_afinfo(afinfo);
		/* Sheit... I remember I did this right. Apparently,
		 * it was magically lost, so this code needs audit */
		x->u.rt6.rt6i_flags    = rt0->rt6i_flags&(RTCF_BROADCAST|RTCF_MULTICAST|RTCF_LOCAL);
		x->u.rt6.rt6i_metric   = rt0->rt6i_metric;
		x->u.rt6.rt6i_node     = rt0->rt6i_node;
		x->u.rt6.rt6i_gateway  = rt0->rt6i_gateway;
		memcpy(&x->u.rt6.rt6i_gateway, &rt0->rt6i_gateway, sizeof(x->u.rt6.rt6i_gateway));
		x->u.rt6.rt6i_dst      = rt0->rt6i_dst;
		x->u.rt6.rt6i_src      = rt0->rt6i_src;
		x->u.rt6.rt6i_idev     = rt0->rt6i_idev;
		in6_dev_hold(rt0->rt6i_idev);
		__xfrm6_bundle_len_dec(&header_len, &nfheader_len, x->u.dst.xfrm);
		trailer_len -= x->u.dst.xfrm->props.trailer_len;
	}

	xfrm_init_pmtu(dst);
	return 0;

error:
	if (dst)
		dst_free(dst);
	return err;
}

static inline void
_decode_session6(struct sk_buff *skb, struct flowi *fl)
{
	u16 offset = skb_network_header_len(skb);
	struct ipv6hdr *hdr = ipv6_hdr(skb);
	struct ipv6_opt_hdr *exthdr;
	const unsigned char *nh = skb_network_header(skb);
	u8 nexthdr = nh[IP6CB(skb)->nhoff];

	memset(fl, 0, sizeof(struct flowi));
	ipv6_addr_copy(&fl->fl6_dst, &hdr->daddr);
	ipv6_addr_copy(&fl->fl6_src, &hdr->saddr);

	while (pskb_may_pull(skb, nh + offset + 1 - skb->data)) {
		nh = skb_network_header(skb);
		exthdr = (struct ipv6_opt_hdr *)(nh + offset);

		switch (nexthdr) {
		case NEXTHDR_ROUTING:
		case NEXTHDR_HOP:
		case NEXTHDR_DEST:
			offset += ipv6_optlen(exthdr);
			nexthdr = exthdr->nexthdr;
			exthdr = (struct ipv6_opt_hdr *)(nh + offset);
			break;

		case IPPROTO_UDP:
		case IPPROTO_UDPLITE:
		case IPPROTO_TCP:
		case IPPROTO_SCTP:
		case IPPROTO_DCCP:
			if (pskb_may_pull(skb, nh + offset + 4 - skb->data)) {
				__be16 *ports = (__be16 *)exthdr;

				fl->fl_ip_sport = ports[0];
				fl->fl_ip_dport = ports[1];
			}
			fl->proto = nexthdr;
			return;

		case IPPROTO_ICMPV6:
			if (pskb_may_pull(skb, nh + offset + 2 - skb->data)) {
				u8 *icmp = (u8 *)exthdr;

				fl->fl_icmp_type = icmp[0];
				fl->fl_icmp_code = icmp[1];
			}
			fl->proto = nexthdr;
			return;

#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE)
		case IPPROTO_MH:
			if (pskb_may_pull(skb, nh + offset + 3 - skb->data)) {
				struct ip6_mh *mh;
				mh = (struct ip6_mh *)exthdr;

				fl->fl_mh_type = mh->ip6mh_type;
			}
			fl->proto = nexthdr;
			return;
#endif

		/* XXX Why are there these headers? */
		case IPPROTO_AH:
		case IPPROTO_ESP:
		case IPPROTO_COMP:
		default:
			fl->fl_ipsec_spi = 0;
			fl->proto = nexthdr;
			return;
		}
	}
}

static inline int xfrm6_garbage_collect(void)
{
	xfrm6_policy_afinfo.garbage_collect();
	return (atomic_read(&xfrm6_dst_ops.entries) > xfrm6_dst_ops.gc_thresh*2);
}

static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu)
{
	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
	struct dst_entry *path = xdst->route;

	path->ops->update_pmtu(path, mtu);
}

static void xfrm6_dst_destroy(struct dst_entry *dst)
{
	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;

	if (likely(xdst->u.rt6.rt6i_idev))
		in6_dev_put(xdst->u.rt6.rt6i_idev);
	xfrm_dst_destroy(xdst);
}

static void xfrm6_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
			     int unregister)
{
	struct xfrm_dst *xdst;

	if (!unregister)
		return;

	xdst = (struct xfrm_dst *)dst;
	if (xdst->u.rt6.rt6i_idev->dev == dev) {
		struct inet6_dev *loopback_idev = in6_dev_get(init_net.loopback_dev);
		BUG_ON(!loopback_idev);

		do {
			in6_dev_put(xdst->u.rt6.rt6i_idev);
			xdst->u.rt6.rt6i_idev = loopback_idev;
			in6_dev_hold(loopback_idev);
			xdst = (struct xfrm_dst *)xdst->u.dst.child;
		} while (xdst->u.dst.xfrm);

		__in6_dev_put(loopback_idev);
	}

	xfrm_dst_ifdown(dst, dev);
}

static struct dst_ops xfrm6_dst_ops = {
	.family =		AF_INET6,
	.protocol =		__constant_htons(ETH_P_IPV6),
	.gc =			xfrm6_garbage_collect,
	.update_pmtu =		xfrm6_update_pmtu,
	.destroy =		xfrm6_dst_destroy,
	.ifdown =		xfrm6_dst_ifdown,
	.gc_thresh =		1024,
	.entry_size =		sizeof(struct xfrm_dst),
};

static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
	.family =		AF_INET6,
	.dst_ops =		&xfrm6_dst_ops,
	.dst_lookup =		xfrm6_dst_lookup,
	.get_saddr = 		xfrm6_get_saddr,
	.find_bundle =		__xfrm6_find_bundle,
	.bundle_create =	__xfrm6_bundle_create,
	.decode_session =	_decode_session6,
};

static void __init xfrm6_policy_init(void)
{
	xfrm_policy_register_afinfo(&xfrm6_policy_afinfo);
}

static void xfrm6_policy_fini(void)
{
	xfrm_policy_unregister_afinfo(&xfrm6_policy_afinfo);
}

void __init xfrm6_init(void)
{
	xfrm6_policy_init();
	xfrm6_state_init();
}

void xfrm6_fini(void)
{
	//xfrm6_input_fini();
	xfrm6_policy_fini();
	xfrm6_state_fini();
}
Commit	Line	Data
1da177e4 LT	1	/*
	2	* xfrm6_policy.c: based on xfrm4_policy.c
	3	*
	4	* Authors:
	5	* Mitsuru KANDA @USAGI
	6	* Kazunori MIYAZAWA @USAGI
	7	* Kunihiro Ishiguro <kunihiro@ipinfusion.com>
	8	* IPv6 support
	9	* YOSHIFUJI Hideaki
	10	* Split up af-specific portion
1ab1457c	11	*
1da177e4 LT	12	*/
1da177e4 LT	13
aabc9761	14	#include <linux/compiler.h>
aabc9761 HX	15	#include <linux/netdevice.h>
aabc9761 HX	16	#include <net/addrconf.h>
1da177e4 LT	17	#include <net/xfrm.h>
	18	#include <net/ip.h>
	19	#include <net/ipv6.h>
	20	#include <net/ip6_route.h>
59fbb3a6	21	#if defined(CONFIG_IPV6_MIP6) \|\| defined(CONFIG_IPV6_MIP6_MODULE)
2ce4272a MN	22	#include <net/mip6.h>
2ce4272a MN	23	#endif
1da177e4 LT	24
	25	static struct dst_ops xfrm6_dst_ops;
	26	static struct xfrm_policy_afinfo xfrm6_policy_afinfo;
	27
4251320f	28	static int xfrm6_dst_lookup(struct xfrm_dst *xdst, struct flowi fl)
1da177e4	29	{
4251320f VN	30	struct dst_entry *dst = ip6_route_output(NULL, fl);
	31	int err = dst->error;
	32	if (!err)
	33	xdst = (struct xfrm_dst ) dst;
	34	else
	35	dst_release(dst);
1da177e4 LT	36	return err;
	37	}
	38
a1e59abf PM	39	static int xfrm6_get_saddr(xfrm_address_t saddr, xfrm_address_t daddr)
	40	{
	41	struct rt6_info *rt;
	42	struct flowi fl_tunnel = {
	43	.nl_u = {
	44	.ip6_u = {
	45	.daddr = (struct in6_addr )&daddr->a6,
	46	},
	47	},
	48	};
	49
	50	if (!xfrm6_dst_lookup((struct xfrm_dst **)&rt, &fl_tunnel)) {
	51	ipv6_get_saddr(&rt->u.dst, (struct in6_addr *)&daddr->a6,
	52	(struct in6_addr *)&saddr->a6);
	53	dst_release(&rt->u.dst);
	54	return 0;
	55	}
	56	return -EHOSTUNREACH;
	57	}
	58
1da177e4 LT	59	static struct dst_entry *
	60	__xfrm6_find_bundle(struct flowi fl, struct xfrm_policy policy)
	61	{
	62	struct dst_entry *dst;
	63
	64	/* Still not clear if we should set fl->fl6_{src,dst}... */
	65	read_lock_bh(&policy->lock);
	66	for (dst = policy->bundles; dst; dst = dst->next) {
	67	struct xfrm_dst xdst = (struct xfrm_dst)dst;
	68	struct in6_addr fl_dst_prefix, fl_src_prefix;
	69
	70	ipv6_addr_prefix(&fl_dst_prefix,
	71	&fl->fl6_dst,
	72	xdst->u.rt6.rt6i_dst.plen);
	73	ipv6_addr_prefix(&fl_src_prefix,
	74	&fl->fl6_src,
	75	xdst->u.rt6.rt6i_src.plen);
	76	if (ipv6_addr_equal(&xdst->u.rt6.rt6i_dst.addr, &fl_dst_prefix) &&
	77	ipv6_addr_equal(&xdst->u.rt6.rt6i_src.addr, &fl_src_prefix) &&
5b368e61	78	xfrm_bundle_ok(policy, xdst, fl, AF_INET6,
e53820de MN	79	(xdst->u.rt6.rt6i_dst.plen != 128 \|\|
e53820de MN	80	xdst->u.rt6.rt6i_src.plen != 128))) {
1da177e4 LT	81	dst_clone(dst);
	82	break;
	83	}
	84	}
	85	read_unlock_bh(&policy->lock);
	86	return dst;
	87	}
	88
99505a84 MN	89	static inline struct in6_addr*
	90	__xfrm6_bundle_addr_remote(struct xfrm_state x, struct in6_addr addr)
	91	{
	92	return (x->type->remote_addr) ?
	93	(struct in6_addr)x->type->remote_addr(x, (xfrm_address_t )addr) :
	94	(struct in6_addr*)&x->id.daddr;
	95	}
	96
	97	static inline struct in6_addr*
	98	__xfrm6_bundle_addr_local(struct xfrm_state x, struct in6_addr addr)
	99	{
	100	return (x->type->local_addr) ?
	101	(struct in6_addr)x->type->local_addr(x, (xfrm_address_t )addr) :
	102	(struct in6_addr*)&x->props.saddr;
	103	}
	104
1b5c2299 MN	105	static inline void
	106	__xfrm6_bundle_len_inc(int len, int nflen, struct xfrm_state *x)
	107	{
	108	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
	109	*nflen += x->props.header_len;
	110	else
	111	*len += x->props.header_len;
	112	}
	113
	114	static inline void
	115	__xfrm6_bundle_len_dec(int len, int nflen, struct xfrm_state *x)
	116	{
	117	if (x->type->flags & XFRM_TYPE_NON_FRAGMENT)
	118	*nflen -= x->props.header_len;
	119	else
	120	*len -= x->props.header_len;
	121	}
	122
1da177e4 LT	123	/* Allocate chain of dst_entry's, attach known xfrm's, calculate
	124	* all the metrics... Shortly, bundle a bundle.
	125	*/
	126
	127	static int
	128	__xfrm6_bundle_create(struct xfrm_policy policy, struct xfrm_state *xfrm, int nx,
	129	struct flowi fl, struct dst_entry *dst_p)
	130	{
	131	struct dst_entry dst, dst_prev;
	132	struct rt6_info rt0 = (struct rt6_info)(*dst_p);
	133	struct rt6_info *rt = rt0;
1da177e4 LT	134	struct flowi fl_tunnel = {
	135	.nl_u = {
	136	.ip6_u = {
c82f963e MK	137	.saddr = fl->fl6_src,
c82f963e MK	138	.daddr = fl->fl6_dst,
1da177e4 LT	139	}
	140	}
	141	};
	142	int i;
	143	int err = 0;
	144	int header_len = 0;
1b5c2299	145	int nfheader_len = 0;
1da177e4 LT	146	int trailer_len = 0;
	147
	148	dst = dst_prev = NULL;
	149	dst_hold(&rt->u.dst);
	150
	151	for (i = 0; i < nx; i++) {
	152	struct dst_entry *dst1 = dst_alloc(&xfrm6_dst_ops);
	153	struct xfrm_dst *xdst;
1da177e4 LT	154
	155	if (unlikely(dst1 == NULL)) {
	156	err = -ENOBUFS;
	157	dst_release(&rt->u.dst);
	158	goto error;
	159	}
	160
	161	if (!dst)
	162	dst = dst1;
	163	else {
	164	dst_prev->child = dst1;
	165	dst1->flags \|= DST_NOHASH;
	166	dst_clone(dst1);
	167	}
	168
	169	xdst = (struct xfrm_dst *)dst1;
	170	xdst->route = &rt->u.dst;
9d4a706d	171	xdst->genid = xfrm[i]->genid;
92d63dec HY	172	if (rt->rt6i_node)
92d63dec HY	173	xdst->route_cookie = rt->rt6i_node->fn_sernum;
1da177e4 LT	174
	175	dst1->next = dst_prev;
	176	dst_prev = dst1;
c82f963e	177
1b5c2299	178	__xfrm6_bundle_len_inc(&header_len, &nfheader_len, xfrm[i]);
1da177e4 LT	179	trailer_len += xfrm[i]->props.trailer_len;
1da177e4 LT	180
bda390d5 MN	181	if (xfrm[i]->props.mode == XFRM_MODE_TUNNEL \|\|
bda390d5 MN	182	xfrm[i]->props.mode == XFRM_MODE_ROUTEOPTIMIZATION) {
c82f963e MK	183	unsigned short encap_family = xfrm[i]->props.family;
	184	switch(encap_family) {
	185	case AF_INET:
	186	fl_tunnel.fl4_dst = xfrm[i]->id.daddr.a4;
	187	fl_tunnel.fl4_src = xfrm[i]->props.saddr.a4;
	188	break;
	189	case AF_INET6:
bda390d5 MN	190	ipv6_addr_copy(&fl_tunnel.fl6_dst, __xfrm6_bundle_addr_remote(xfrm[i], &fl->fl6_dst));
bda390d5 MN	191
d3f23dfe	192	ipv6_addr_copy(&fl_tunnel.fl6_src, __xfrm6_bundle_addr_local(xfrm[i], &fl->fl6_src));
c82f963e MK	193	break;
	194	default:
	195	BUG_ON(1);
	196	}
	197
1da177e4	198	err = xfrm_dst_lookup((struct xfrm_dst **) &rt,
c82f963e	199	&fl_tunnel, encap_family);
1da177e4 LT	200	if (err)
	201	goto error;
	202	} else
	203	dst_hold(&rt->u.dst);
	204	}
	205
	206	dst_prev->child = &rt->u.dst;
	207	dst->path = &rt->u.dst;
92d63dec HY	208	if (rt->rt6i_node)
92d63dec HY	209	((struct xfrm_dst *)dst)->path_cookie = rt->rt6i_node->fn_sernum;
1da177e4 LT	210
	211	*dst_p = dst;
	212	dst = dst_prev;
	213
	214	dst_prev = *dst_p;
	215	i = 0;
	216	for (; dst_prev != &rt->u.dst; dst_prev = dst_prev->child) {
	217	struct xfrm_dst x = (struct xfrm_dst)dst_prev;
c82f963e	218	struct xfrm_state_afinfo *afinfo;
1da177e4 LT	219
	220	dst_prev->xfrm = xfrm[i++];
	221	dst_prev->dev = rt->u.dst.dev;
	222	if (rt->u.dst.dev)
	223	dev_hold(rt->u.dst.dev);
	224	dst_prev->obsolete = -1;
	225	dst_prev->flags \|= DST_HOST;
	226	dst_prev->lastuse = jiffies;
	227	dst_prev->header_len = header_len;
1b5c2299	228	dst_prev->nfheader_len = nfheader_len;
1da177e4 LT	229	dst_prev->trailer_len = trailer_len;
	230	memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
	231
	232	/* Copy neighbour for reachability confirmation */
	233	dst_prev->neighbour = neigh_clone(rt->u.dst.neighbour);
	234	dst_prev->input = rt->u.dst.input;
c82f963e MK	235	/* XXX: When IPv4 is implemented as module and can be unloaded,
	236	* we should manage reference to xfrm4_output in afinfo->output.
	237	* Miyazawa
	238	*/
	239	afinfo = xfrm_state_get_afinfo(dst_prev->xfrm->props.family);
	240	if (!afinfo) {
	241	dst = *dst_p;
	242	goto error;
3ff50b79 SH	243	}
3ff50b79 SH	244
c82f963e MK	245	dst_prev->output = afinfo->output;
c82f963e MK	246	xfrm_state_put_afinfo(afinfo);
1da177e4 LT	247	/* Sheit... I remember I did this right. Apparently,
	248	* it was magically lost, so this code needs audit */
	249	x->u.rt6.rt6i_flags = rt0->rt6i_flags&(RTCF_BROADCAST\|RTCF_MULTICAST\|RTCF_LOCAL);
	250	x->u.rt6.rt6i_metric = rt0->rt6i_metric;
	251	x->u.rt6.rt6i_node = rt0->rt6i_node;
	252	x->u.rt6.rt6i_gateway = rt0->rt6i_gateway;
1ab1457c	253	memcpy(&x->u.rt6.rt6i_gateway, &rt0->rt6i_gateway, sizeof(x->u.rt6.rt6i_gateway));
1da177e4	254	x->u.rt6.rt6i_dst = rt0->rt6i_dst;
1ab1457c	255	x->u.rt6.rt6i_src = rt0->rt6i_src;
aabc9761 HX	256	x->u.rt6.rt6i_idev = rt0->rt6i_idev;
aabc9761 HX	257	in6_dev_hold(rt0->rt6i_idev);
1b5c2299	258	__xfrm6_bundle_len_dec(&header_len, &nfheader_len, x->u.dst.xfrm);
1da177e4 LT	259	trailer_len -= x->u.dst.xfrm->props.trailer_len;
	260	}
	261
	262	xfrm_init_pmtu(dst);
	263	return 0;
	264
	265	error:
	266	if (dst)
	267	dst_free(dst);
	268	return err;
	269	}
	270
	271	static inline void
	272	_decode_session6(struct sk_buff skb, struct flowi fl)
	273	{
cfe1fc77	274	u16 offset = skb_network_header_len(skb);
0660e03f	275	struct ipv6hdr *hdr = ipv6_hdr(skb);
e3cae904	276	struct ipv6_opt_hdr *exthdr;
d56f90a7 ACM	277	const unsigned char *nh = skb_network_header(skb);
d56f90a7 ACM	278	u8 nexthdr = nh[IP6CB(skb)->nhoff];
1da177e4 LT	279
	280	memset(fl, 0, sizeof(struct flowi));
	281	ipv6_addr_copy(&fl->fl6_dst, &hdr->daddr);
	282	ipv6_addr_copy(&fl->fl6_src, &hdr->saddr);
	283
d56f90a7 ACM	284	while (pskb_may_pull(skb, nh + offset + 1 - skb->data)) {
	285	nh = skb_network_header(skb);
	286	exthdr = (struct ipv6_opt_hdr *)(nh + offset);
e3cae904	287
1da177e4 LT	288	switch (nexthdr) {
	289	case NEXTHDR_ROUTING:
	290	case NEXTHDR_HOP:
	291	case NEXTHDR_DEST:
	292	offset += ipv6_optlen(exthdr);
	293	nexthdr = exthdr->nexthdr;
d56f90a7	294	exthdr = (struct ipv6_opt_hdr *)(nh + offset);
1da177e4 LT	295	break;
	296
	297	case IPPROTO_UDP:
ba4e58ec	298	case IPPROTO_UDPLITE:
1da177e4 LT	299	case IPPROTO_TCP:
1da177e4 LT	300	case IPPROTO_SCTP:
9e999993	301	case IPPROTO_DCCP:
d56f90a7	302	if (pskb_may_pull(skb, nh + offset + 4 - skb->data)) {
8c689a6e	303	__be16 ports = (__be16 )exthdr;
1da177e4 LT	304
	305	fl->fl_ip_sport = ports[0];
	306	fl->fl_ip_dport = ports[1];
	307	}
	308	fl->proto = nexthdr;
	309	return;
	310
	311	case IPPROTO_ICMPV6:
d56f90a7	312	if (pskb_may_pull(skb, nh + offset + 2 - skb->data)) {
1da177e4 LT	313	u8 icmp = (u8 )exthdr;
	314
	315	fl->fl_icmp_type = icmp[0];
	316	fl->fl_icmp_code = icmp[1];
	317	}
	318	fl->proto = nexthdr;
	319	return;
	320
59fbb3a6	321	#if defined(CONFIG_IPV6_MIP6) \|\| defined(CONFIG_IPV6_MIP6_MODULE)
2ce4272a	322	case IPPROTO_MH:
d56f90a7	323	if (pskb_may_pull(skb, nh + offset + 3 - skb->data)) {
2ce4272a MN	324	struct ip6_mh *mh;
	325	mh = (struct ip6_mh *)exthdr;
	326
	327	fl->fl_mh_type = mh->ip6mh_type;
	328	}
	329	fl->proto = nexthdr;
	330	return;
	331	#endif
	332
1da177e4 LT	333	/* XXX Why are there these headers? */
	334	case IPPROTO_AH:
	335	case IPPROTO_ESP:
	336	case IPPROTO_COMP:
	337	default:
	338	fl->fl_ipsec_spi = 0;
	339	fl->proto = nexthdr;
	340	return;
3ff50b79	341	}
1da177e4 LT	342	}
	343	}
	344
	345	static inline int xfrm6_garbage_collect(void)
	346	{
1da177e4	347	xfrm6_policy_afinfo.garbage_collect();
1da177e4 LT	348	return (atomic_read(&xfrm6_dst_ops.entries) > xfrm6_dst_ops.gc_thresh*2);
	349	}
	350
	351	static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu)
	352	{
	353	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	354	struct dst_entry *path = xdst->route;
	355
	356	path->ops->update_pmtu(path, mtu);
	357	}
	358
aabc9761 HX	359	static void xfrm6_dst_destroy(struct dst_entry *dst)
	360	{
	361	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	362
	363	if (likely(xdst->u.rt6.rt6i_idev))
	364	in6_dev_put(xdst->u.rt6.rt6i_idev);
	365	xfrm_dst_destroy(xdst);
	366	}
	367
	368	static void xfrm6_dst_ifdown(struct dst_entry dst, struct net_device dev,
	369	int unregister)
	370	{
	371	struct xfrm_dst *xdst;
	372
	373	if (!unregister)
	374	return;
	375
	376	xdst = (struct xfrm_dst *)dst;
	377	if (xdst->u.rt6.rt6i_idev->dev == dev) {
2774c7ab	378	struct inet6_dev *loopback_idev = in6_dev_get(init_net.loopback_dev);
aabc9761 HX	379	BUG_ON(!loopback_idev);
	380
	381	do {
	382	in6_dev_put(xdst->u.rt6.rt6i_idev);
	383	xdst->u.rt6.rt6i_idev = loopback_idev;
	384	in6_dev_hold(loopback_idev);
	385	xdst = (struct xfrm_dst *)xdst->u.dst.child;
	386	} while (xdst->u.dst.xfrm);
	387
	388	__in6_dev_put(loopback_idev);
	389	}
	390
	391	xfrm_dst_ifdown(dst, dev);
	392	}
	393
1da177e4 LT	394	static struct dst_ops xfrm6_dst_ops = {
	395	.family = AF_INET6,
	396	.protocol = __constant_htons(ETH_P_IPV6),
	397	.gc = xfrm6_garbage_collect,
	398	.update_pmtu = xfrm6_update_pmtu,
aabc9761 HX	399	.destroy = xfrm6_dst_destroy,
aabc9761 HX	400	.ifdown = xfrm6_dst_ifdown,
1da177e4 LT	401	.gc_thresh = 1024,
	402	.entry_size = sizeof(struct xfrm_dst),
	403	};
	404
	405	static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
	406	.family = AF_INET6,
1da177e4 LT	407	.dst_ops = &xfrm6_dst_ops,
1da177e4 LT	408	.dst_lookup = xfrm6_dst_lookup,
a1e59abf	409	.get_saddr = xfrm6_get_saddr,
1da177e4 LT	410	.find_bundle = __xfrm6_find_bundle,
	411	.bundle_create = __xfrm6_bundle_create,
	412	.decode_session = _decode_session6,
	413	};
	414
	415	static void __init xfrm6_policy_init(void)
	416	{
	417	xfrm_policy_register_afinfo(&xfrm6_policy_afinfo);
	418	}
	419
	420	static void xfrm6_policy_fini(void)
	421	{
	422	xfrm_policy_unregister_afinfo(&xfrm6_policy_afinfo);
	423	}
	424
	425	void __init xfrm6_init(void)
	426	{
	427	xfrm6_policy_init();
	428	xfrm6_state_init();
	429	}
	430
	431	void xfrm6_fini(void)
	432	{
	433	//xfrm6_input_fini();
	434	xfrm6_policy_fini();
	435	xfrm6_state_fini();
	436	}