[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/err.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <net/addrconf.h>
#include <net/dst.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 xfrm_policy_afinfo xfrm6_policy_afinfo;

static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos,
					  xfrm_address_t *saddr,
					  xfrm_address_t *daddr)
{
	struct flowi fl = {};
	struct dst_entry *dst;
	int err;

	memcpy(&fl.fl6_dst, daddr, sizeof(fl.fl6_dst));
	if (saddr)
		memcpy(&fl.fl6_src, saddr, sizeof(fl.fl6_src));

	dst = ip6_route_output(net, NULL, &fl);

	err = dst->error;
	if (dst->error) {
		dst_release(dst);
		dst = ERR_PTR(err);
	}

	return dst;
}

static int xfrm6_get_saddr(struct net *net,
			   xfrm_address_t *saddr, xfrm_address_t *daddr)
{
	struct dst_entry *dst;
	struct net_device *dev;

	dst = xfrm6_dst_lookup(net, 0, NULL, daddr);
	if (IS_ERR(dst))
		return -EHOSTUNREACH;

	dev = ip6_dst_idev(dst)->dev;
	ipv6_dev_get_saddr(dev_net(dev), dev,
			   (struct in6_addr *)&daddr->a6, 0,
			   (struct in6_addr *)&saddr->a6);
	dst_release(dst);
	return 0;
}

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 int xfrm6_get_tos(struct flowi *fl)
{
	return 0;
}

static int xfrm6_init_path(struct xfrm_dst *path, struct dst_entry *dst,
			   int nfheader_len)
{
	if (dst->ops->family == AF_INET6) {
		struct rt6_info *rt = (struct rt6_info*)dst;
		if (rt->rt6i_node)
			path->path_cookie = rt->rt6i_node->fn_sernum;
	}

	path->u.rt6.rt6i_nfheader_len = nfheader_len;

	return 0;
}

static int xfrm6_fill_dst(struct xfrm_dst *xdst, struct net_device *dev,
			  struct flowi *fl)
{
	struct rt6_info *rt = (struct rt6_info*)xdst->route;

	xdst->u.dst.dev = dev;
	dev_hold(dev);

	xdst->u.rt6.rt6i_idev = in6_dev_get(rt->u.dst.dev);
	if (!xdst->u.rt6.rt6i_idev)
		return -ENODEV;

	/* Sheit... I remember I did this right. Apparently,
	 * it was magically lost, so this code needs audit */
	xdst->u.rt6.rt6i_flags = rt->rt6i_flags & (RTF_ANYCAST |
						   RTF_LOCAL);
	xdst->u.rt6.rt6i_metric = rt->rt6i_metric;
	xdst->u.rt6.rt6i_node = rt->rt6i_node;
	if (rt->rt6i_node)
		xdst->route_cookie = rt->rt6i_node->fn_sernum;
	xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway;
	xdst->u.rt6.rt6i_dst = rt->rt6i_dst;
	xdst->u.rt6.rt6i_src = rt->rt6i_src;

	return 0;
}

static inline void
_decode_session6(struct sk_buff *skb, struct flowi *fl, int reverse)
{
	int onlyproto = 0;
	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, reverse ? &hdr->saddr : &hdr->daddr);
	ipv6_addr_copy(&fl->fl6_src, reverse ? &hdr->daddr : &hdr->saddr);

	while (nh + offset + 1 < skb->data ||
	       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_FRAGMENT:
			onlyproto = 1;
		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 (!onlyproto && (nh + offset + 4 < skb->data ||
			     pskb_may_pull(skb, nh + offset + 4 - skb->data))) {
				__be16 *ports = (__be16 *)exthdr;

				fl->fl_ip_sport = ports[!!reverse];
				fl->fl_ip_dport = ports[!reverse];
			}
			fl->proto = nexthdr;
			return;

		case IPPROTO_ICMPV6:
			if (!onlyproto && 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 (!onlyproto && 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(struct dst_ops *ops)
{
	struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops);

	xfrm6_policy_afinfo.garbage_collect(net);
	return (atomic_read(&ops->entries) > 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(dev_net(dev)->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 =		cpu_to_be16(ETH_P_IPV6),
	.gc =			xfrm6_garbage_collect,
	.update_pmtu =		xfrm6_update_pmtu,
	.destroy =		xfrm6_dst_destroy,
	.ifdown =		xfrm6_dst_ifdown,
	.local_out =		__ip6_local_out,
	.gc_thresh =		1024,
	.entries =		ATOMIC_INIT(0),
};

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,
	.decode_session =	_decode_session6,
	.get_tos =		xfrm6_get_tos,
	.init_path =		xfrm6_init_path,
	.fill_dst =		xfrm6_fill_dst,
};

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

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

#ifdef CONFIG_SYSCTL
static struct ctl_table xfrm6_policy_table[] = {
	{
		.procname       = "xfrm6_gc_thresh",
		.data	   	= &init_net.xfrm.xfrm6_dst_ops.gc_thresh,
		.maxlen	 	= sizeof(int),
		.mode	   	= 0644,
		.proc_handler   = proc_dointvec,
	},
	{ }
};

static struct ctl_table_header *sysctl_hdr;
#endif

int __init xfrm6_init(void)
{
	int ret;
	unsigned int gc_thresh;

	/*
	 * We need a good default value for the xfrm6 gc threshold.
	 * In ipv4 we set it to the route hash table size * 8, which
	 * is half the size of the maximaum route cache for ipv4.  It
	 * would be good to do the same thing for v6, except the table is
	 * constructed differently here.  Here each table for a net namespace
	 * can have FIB_TABLE_HASHSZ entries, so lets go with the same
	 * computation that we used for ipv4 here.  Also, lets keep the initial
	 * gc_thresh to a minimum of 1024, since, the ipv6 route cache defaults
	 * to that as a minimum as well
	 */
	gc_thresh = FIB6_TABLE_HASHSZ * 8;
	xfrm6_dst_ops.gc_thresh = (gc_thresh < 1024) ? 1024 : gc_thresh;

	ret = xfrm6_policy_init();
	if (ret)
		goto out;

	ret = xfrm6_state_init();
	if (ret)
		goto out_policy;

#ifdef CONFIG_SYSCTL
	sysctl_hdr = register_net_sysctl_table(&init_net, net_ipv6_ctl_path,
						xfrm6_policy_table);
#endif
out:
	return ret;
out_policy:
	xfrm6_policy_fini();
	goto out;
}

void xfrm6_fini(void)
{
#ifdef CONFIG_SYSCTL
	if (sysctl_hdr)
		unregister_net_sysctl_table(sysctl_hdr);
#endif
	//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
66cdb3ca HX	14	#include <linux/err.h>
66cdb3ca HX	15	#include <linux/kernel.h>
aabc9761 HX	16	#include <linux/netdevice.h>
aabc9761 HX	17	#include <net/addrconf.h>
45ff5a3f	18	#include <net/dst.h>
1da177e4 LT	19	#include <net/xfrm.h>
	20	#include <net/ip.h>
	21	#include <net/ipv6.h>
	22	#include <net/ip6_route.h>
59fbb3a6	23	#if defined(CONFIG_IPV6_MIP6) \|\| defined(CONFIG_IPV6_MIP6_MODULE)
2ce4272a MN	24	#include <net/mip6.h>
2ce4272a MN	25	#endif
1da177e4	26
1da177e4 LT	27	static struct xfrm_policy_afinfo xfrm6_policy_afinfo;
1da177e4 LT	28
c5b3cf46 AD	29	static struct dst_entry xfrm6_dst_lookup(struct net net, int tos,
c5b3cf46 AD	30	xfrm_address_t *saddr,
66cdb3ca	31	xfrm_address_t *daddr)
1da177e4	32	{
66cdb3ca HX	33	struct flowi fl = {};
	34	struct dst_entry *dst;
	35	int err;
	36
	37	memcpy(&fl.fl6_dst, daddr, sizeof(fl.fl6_dst));
	38	if (saddr)
	39	memcpy(&fl.fl6_src, saddr, sizeof(fl.fl6_src));
	40
c5b3cf46	41	dst = ip6_route_output(net, NULL, &fl);
66cdb3ca HX	42
	43	err = dst->error;
	44	if (dst->error) {
4251320f	45	dst_release(dst);
66cdb3ca HX	46	dst = ERR_PTR(err);
	47	}
	48
	49	return dst;
1da177e4 LT	50	}
1da177e4 LT	51
fbda33b2 AD	52	static int xfrm6_get_saddr(struct net *net,
fbda33b2 AD	53	xfrm_address_t saddr, xfrm_address_t daddr)
a1e59abf	54	{
66cdb3ca	55	struct dst_entry *dst;
191cd582	56	struct net_device *dev;
66cdb3ca	57
fbda33b2	58	dst = xfrm6_dst_lookup(net, 0, NULL, daddr);
66cdb3ca HX	59	if (IS_ERR(dst))
	60	return -EHOSTUNREACH;
	61
191cd582 BH	62	dev = ip6_dst_idev(dst)->dev;
191cd582 BH	63	ipv6_dev_get_saddr(dev_net(dev), dev,
7cbca67c	64	(struct in6_addr *)&daddr->a6, 0,
5e5f3f0f	65	(struct in6_addr *)&saddr->a6);
66cdb3ca HX	66	dst_release(dst);
66cdb3ca HX	67	return 0;
a1e59abf PM	68	}
a1e59abf PM	69
1da177e4 LT	70	static struct dst_entry *
	71	__xfrm6_find_bundle(struct flowi fl, struct xfrm_policy policy)
	72	{
	73	struct dst_entry *dst;
	74
	75	/* Still not clear if we should set fl->fl6_{src,dst}... */
	76	read_lock_bh(&policy->lock);
	77	for (dst = policy->bundles; dst; dst = dst->next) {
	78	struct xfrm_dst xdst = (struct xfrm_dst)dst;
	79	struct in6_addr fl_dst_prefix, fl_src_prefix;
	80
	81	ipv6_addr_prefix(&fl_dst_prefix,
	82	&fl->fl6_dst,
	83	xdst->u.rt6.rt6i_dst.plen);
	84	ipv6_addr_prefix(&fl_src_prefix,
	85	&fl->fl6_src,
	86	xdst->u.rt6.rt6i_src.plen);
	87	if (ipv6_addr_equal(&xdst->u.rt6.rt6i_dst.addr, &fl_dst_prefix) &&
	88	ipv6_addr_equal(&xdst->u.rt6.rt6i_src.addr, &fl_src_prefix) &&
5b368e61	89	xfrm_bundle_ok(policy, xdst, fl, AF_INET6,
e53820de MN	90	(xdst->u.rt6.rt6i_dst.plen != 128 \|\|
e53820de MN	91	xdst->u.rt6.rt6i_src.plen != 128))) {
1da177e4 LT	92	dst_clone(dst);
	93	break;
	94	}
	95	}
	96	read_unlock_bh(&policy->lock);
	97	return dst;
	98	}
	99
25ee3286	100	static int xfrm6_get_tos(struct flowi *fl)
1da177e4	101	{
25ee3286 HX	102	return 0;
25ee3286 HX	103	}
1da177e4	104
a1b05140 MN	105	static int xfrm6_init_path(struct xfrm_dst path, struct dst_entry dst,
	106	int nfheader_len)
	107	{
	108	if (dst->ops->family == AF_INET6) {
	109	struct rt6_info rt = (struct rt6_info)dst;
	110	if (rt->rt6i_node)
	111	path->path_cookie = rt->rt6i_node->fn_sernum;
	112	}
	113
	114	path->u.rt6.rt6i_nfheader_len = nfheader_len;
	115
	116	return 0;
	117	}
	118
87c1e12b HX	119	static int xfrm6_fill_dst(struct xfrm_dst xdst, struct net_device dev,
87c1e12b HX	120	struct flowi *fl)
25ee3286 HX	121	{
25ee3286 HX	122	struct rt6_info rt = (struct rt6_info)xdst->route;
1da177e4	123
25ee3286 HX	124	xdst->u.dst.dev = dev;
25ee3286 HX	125	dev_hold(dev);
1da177e4	126
25ee3286 HX	127	xdst->u.rt6.rt6i_idev = in6_dev_get(rt->u.dst.dev);
	128	if (!xdst->u.rt6.rt6i_idev)
	129	return -ENODEV;
1da177e4	130
25ee3286 HX	131	/* Sheit... I remember I did this right. Apparently,
	132	* it was magically lost, so this code needs audit */
	133	xdst->u.rt6.rt6i_flags = rt->rt6i_flags & (RTF_ANYCAST \|
	134	RTF_LOCAL);
	135	xdst->u.rt6.rt6i_metric = rt->rt6i_metric;
	136	xdst->u.rt6.rt6i_node = rt->rt6i_node;
a1b05140 MN	137	if (rt->rt6i_node)
a1b05140 MN	138	xdst->route_cookie = rt->rt6i_node->fn_sernum;
25ee3286 HX	139	xdst->u.rt6.rt6i_gateway = rt->rt6i_gateway;
	140	xdst->u.rt6.rt6i_dst = rt->rt6i_dst;
	141	xdst->u.rt6.rt6i_src = rt->rt6i_src;
1da177e4	142
1da177e4	143	return 0;
1da177e4 LT	144	}
	145
	146	static inline void
d5422efe	147	_decode_session6(struct sk_buff skb, struct flowi fl, int reverse)
1da177e4	148	{
7e3a42a1	149	int onlyproto = 0;
cfe1fc77	150	u16 offset = skb_network_header_len(skb);
0660e03f	151	struct ipv6hdr *hdr = ipv6_hdr(skb);
e3cae904	152	struct ipv6_opt_hdr *exthdr;
d56f90a7 ACM	153	const unsigned char *nh = skb_network_header(skb);
d56f90a7 ACM	154	u8 nexthdr = nh[IP6CB(skb)->nhoff];
1da177e4 LT	155
1da177e4 LT	156	memset(fl, 0, sizeof(struct flowi));
d5422efe HX	157	ipv6_addr_copy(&fl->fl6_dst, reverse ? &hdr->saddr : &hdr->daddr);
d5422efe HX	158	ipv6_addr_copy(&fl->fl6_src, reverse ? &hdr->daddr : &hdr->saddr);
1da177e4	159
59cae009 WY	160	while (nh + offset + 1 < skb->data \|\|
59cae009 WY	161	pskb_may_pull(skb, nh + offset + 1 - skb->data)) {
d56f90a7 ACM	162	nh = skb_network_header(skb);
d56f90a7 ACM	163	exthdr = (struct ipv6_opt_hdr *)(nh + offset);
e3cae904	164
1da177e4	165	switch (nexthdr) {
7e3a42a1 ND	166	case NEXTHDR_FRAGMENT:
7e3a42a1 ND	167	onlyproto = 1;
1da177e4 LT	168	case NEXTHDR_ROUTING:
	169	case NEXTHDR_HOP:
	170	case NEXTHDR_DEST:
	171	offset += ipv6_optlen(exthdr);
	172	nexthdr = exthdr->nexthdr;
d56f90a7	173	exthdr = (struct ipv6_opt_hdr *)(nh + offset);
1da177e4 LT	174	break;
	175
	176	case IPPROTO_UDP:
ba4e58ec	177	case IPPROTO_UDPLITE:
1da177e4 LT	178	case IPPROTO_TCP:
1da177e4 LT	179	case IPPROTO_SCTP:
9e999993	180	case IPPROTO_DCCP:
59cae009 WY	181	if (!onlyproto && (nh + offset + 4 < skb->data \|\|
59cae009 WY	182	pskb_may_pull(skb, nh + offset + 4 - skb->data))) {
8c689a6e	183	__be16 ports = (__be16 )exthdr;
1da177e4	184
d5422efe HX	185	fl->fl_ip_sport = ports[!!reverse];
d5422efe HX	186	fl->fl_ip_dport = ports[!reverse];
1da177e4 LT	187	}
	188	fl->proto = nexthdr;
	189	return;
	190
	191	case IPPROTO_ICMPV6:
7e3a42a1	192	if (!onlyproto && pskb_may_pull(skb, nh + offset + 2 - skb->data)) {
1da177e4 LT	193	u8 icmp = (u8 )exthdr;
	194
	195	fl->fl_icmp_type = icmp[0];
	196	fl->fl_icmp_code = icmp[1];
	197	}
	198	fl->proto = nexthdr;
	199	return;
	200
59fbb3a6	201	#if defined(CONFIG_IPV6_MIP6) \|\| defined(CONFIG_IPV6_MIP6_MODULE)
2ce4272a	202	case IPPROTO_MH:
7e3a42a1	203	if (!onlyproto && pskb_may_pull(skb, nh + offset + 3 - skb->data)) {
2ce4272a MN	204	struct ip6_mh *mh;
	205	mh = (struct ip6_mh *)exthdr;
	206
	207	fl->fl_mh_type = mh->ip6mh_type;
	208	}
	209	fl->proto = nexthdr;
	210	return;
	211	#endif
	212
1da177e4 LT	213	/* XXX Why are there these headers? */
	214	case IPPROTO_AH:
	215	case IPPROTO_ESP:
	216	case IPPROTO_COMP:
	217	default:
	218	fl->fl_ipsec_spi = 0;
	219	fl->proto = nexthdr;
	220	return;
3ff50b79	221	}
1da177e4 LT	222	}
	223	}
	224
569d3645	225	static inline int xfrm6_garbage_collect(struct dst_ops *ops)
1da177e4	226	{
d7c7544c AD	227	struct net *net = container_of(ops, struct net, xfrm.xfrm6_dst_ops);
	228
	229	xfrm6_policy_afinfo.garbage_collect(net);
	230	return (atomic_read(&ops->entries) > ops->gc_thresh * 2);
1da177e4 LT	231	}
	232
	233	static void xfrm6_update_pmtu(struct dst_entry *dst, u32 mtu)
	234	{
	235	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	236	struct dst_entry *path = xdst->route;
	237
	238	path->ops->update_pmtu(path, mtu);
	239	}
	240
aabc9761 HX	241	static void xfrm6_dst_destroy(struct dst_entry *dst)
	242	{
	243	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	244
	245	if (likely(xdst->u.rt6.rt6i_idev))
	246	in6_dev_put(xdst->u.rt6.rt6i_idev);
	247	xfrm_dst_destroy(xdst);
	248	}
	249
	250	static void xfrm6_dst_ifdown(struct dst_entry dst, struct net_device dev,
	251	int unregister)
	252	{
	253	struct xfrm_dst *xdst;
	254
	255	if (!unregister)
	256	return;
	257
	258	xdst = (struct xfrm_dst *)dst;
	259	if (xdst->u.rt6.rt6i_idev->dev == dev) {
5a3e55d6	260	struct inet6_dev *loopback_idev =
c346dca1	261	in6_dev_get(dev_net(dev)->loopback_dev);
aabc9761 HX	262	BUG_ON(!loopback_idev);
	263
	264	do {
	265	in6_dev_put(xdst->u.rt6.rt6i_idev);
	266	xdst->u.rt6.rt6i_idev = loopback_idev;
	267	in6_dev_hold(loopback_idev);
	268	xdst = (struct xfrm_dst *)xdst->u.dst.child;
	269	} while (xdst->u.dst.xfrm);
	270
	271	__in6_dev_put(loopback_idev);
	272	}
	273
	274	xfrm_dst_ifdown(dst, dev);
	275	}
	276
1da177e4 LT	277	static struct dst_ops xfrm6_dst_ops = {
1da177e4 LT	278	.family = AF_INET6,
09640e63	279	.protocol = cpu_to_be16(ETH_P_IPV6),
1da177e4 LT	280	.gc = xfrm6_garbage_collect,
1da177e4 LT	281	.update_pmtu = xfrm6_update_pmtu,
aabc9761 HX	282	.destroy = xfrm6_dst_destroy,
aabc9761 HX	283	.ifdown = xfrm6_dst_ifdown,
862b82c6	284	.local_out = __ip6_local_out,
1da177e4	285	.gc_thresh = 1024,
e2422970	286	.entries = ATOMIC_INIT(0),
1da177e4 LT	287	};
	288
	289	static struct xfrm_policy_afinfo xfrm6_policy_afinfo = {
	290	.family = AF_INET6,
1da177e4 LT	291	.dst_ops = &xfrm6_dst_ops,
1da177e4 LT	292	.dst_lookup = xfrm6_dst_lookup,
a1e59abf	293	.get_saddr = xfrm6_get_saddr,
1da177e4	294	.find_bundle = __xfrm6_find_bundle,
1da177e4	295	.decode_session = _decode_session6,
25ee3286	296	.get_tos = xfrm6_get_tos,
a1b05140	297	.init_path = xfrm6_init_path,
25ee3286	298	.fill_dst = xfrm6_fill_dst,
1da177e4 LT	299	};
1da177e4 LT	300
0013caba	301	static int __init xfrm6_policy_init(void)
1da177e4	302	{
0013caba	303	return xfrm_policy_register_afinfo(&xfrm6_policy_afinfo);
1da177e4 LT	304	}
	305
	306	static void xfrm6_policy_fini(void)
	307	{
	308	xfrm_policy_unregister_afinfo(&xfrm6_policy_afinfo);
	309	}
	310
db71789c	311	#ifdef CONFIG_SYSCTL
a44a4a00 NH	312	static struct ctl_table xfrm6_policy_table[] = {
a44a4a00 NH	313	{
a44a4a00	314	.procname = "xfrm6_gc_thresh",
d7c7544c	315	.data = &init_net.xfrm.xfrm6_dst_ops.gc_thresh,
a44a4a00 NH	316	.maxlen = sizeof(int),
	317	.mode = 0644,
	318	.proc_handler = proc_dointvec,
	319	},
	320	{ }
	321	};
	322
	323	static struct ctl_table_header *sysctl_hdr;
db71789c	324	#endif
a44a4a00	325
0013caba	326	int __init xfrm6_init(void)
1da177e4	327	{
0013caba	328	int ret;
a33bc5c1	329	unsigned int gc_thresh;
0013caba	330
a33bc5c1 NH	331	/*
	332	* We need a good default value for the xfrm6 gc threshold.
	333	* In ipv4 we set it to the route hash table size * 8, which
	334	* is half the size of the maximaum route cache for ipv4. It
	335	* would be good to do the same thing for v6, except the table is
	336	* constructed differently here. Here each table for a net namespace
	337	* can have FIB_TABLE_HASHSZ entries, so lets go with the same
	338	* computation that we used for ipv4 here. Also, lets keep the initial
	339	* gc_thresh to a minimum of 1024, since, the ipv6 route cache defaults
	340	* to that as a minimum as well
	341	*/
	342	gc_thresh = FIB6_TABLE_HASHSZ * 8;
	343	xfrm6_dst_ops.gc_thresh = (gc_thresh < 1024) ? 1024 : gc_thresh;
d7c7544c AD	344
	345	ret = xfrm6_policy_init();
	346	if (ret)
	347	goto out;
	348
	349	ret = xfrm6_state_init();
	350	if (ret)
	351	goto out_policy;
	352
db71789c	353	#ifdef CONFIG_SYSCTL
a44a4a00 NH	354	sysctl_hdr = register_net_sysctl_table(&init_net, net_ipv6_ctl_path,
a44a4a00 NH	355	xfrm6_policy_table);
db71789c	356	#endif
0013caba DL	357	out:
	358	return ret;
	359	out_policy:
	360	xfrm6_policy_fini();
	361	goto out;
1da177e4 LT	362	}
	363
	364	void xfrm6_fini(void)
	365	{
db71789c	366	#ifdef CONFIG_SYSCTL
a44a4a00 NH	367	if (sysctl_hdr)
a44a4a00 NH	368	unregister_net_sysctl_table(sysctl_hdr);
db71789c	369	#endif
1da177e4 LT	370	//xfrm6_input_fini();
	371	xfrm6_policy_fini();
	372	xfrm6_state_fini();
	373	}