[net-next-2.6.git] / net / ipv4 / xfrm4_policy.c

/*
 * xfrm4_policy.c
 *
 * Changes:
 *	Kazunori MIYAZAWA @USAGI
 * 	YOSHIFUJI Hideaki @USAGI
 *		Split up af-specific portion
 *
 */

#include <linux/err.h>
#include <linux/kernel.h>
#include <linux/inetdevice.h>
#include <net/dst.h>
#include <net/xfrm.h>
#include <net/ip.h>

static struct dst_ops xfrm4_dst_ops;
static struct xfrm_policy_afinfo xfrm4_policy_afinfo;

static struct dst_entry *xfrm4_dst_lookup(int tos, xfrm_address_t *saddr,
					  xfrm_address_t *daddr)
{
	struct flowi fl = {
		.nl_u = {
			.ip4_u = {
				.tos = tos,
				.daddr = daddr->a4,
			},
		},
	};
	struct dst_entry *dst;
	struct rtable *rt;
	int err;

	if (saddr)
		fl.fl4_src = saddr->a4;

	err = __ip_route_output_key(&init_net, &rt, &fl);
	dst = &rt->u.dst;
	if (err)
		dst = ERR_PTR(err);
	return dst;
}

static int xfrm4_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr)
{
	struct dst_entry *dst;
	struct rtable *rt;

	dst = xfrm4_dst_lookup(0, NULL, daddr);
	if (IS_ERR(dst))
		return -EHOSTUNREACH;

	rt = (struct rtable *)dst;
	saddr->a4 = rt->rt_src;
	dst_release(dst);
	return 0;
}

static struct dst_entry *
__xfrm4_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
{
	struct dst_entry *dst;

	read_lock_bh(&policy->lock);
	for (dst = policy->bundles; dst; dst = dst->next) {
		struct xfrm_dst *xdst = (struct xfrm_dst*)dst;
		if (xdst->u.rt.fl.oif == fl->oif &&	/*XXX*/
		    xdst->u.rt.fl.fl4_dst == fl->fl4_dst &&
		    xdst->u.rt.fl.fl4_src == fl->fl4_src &&
		    xdst->u.rt.fl.fl4_tos == fl->fl4_tos &&
		    xfrm_bundle_ok(policy, xdst, fl, AF_INET, 0)) {
			dst_clone(dst);
			break;
		}
	}
	read_unlock_bh(&policy->lock);
	return dst;
}

static int xfrm4_get_tos(struct flowi *fl)
{
	return fl->fl4_tos;
}

static int xfrm4_init_path(struct xfrm_dst *path, struct dst_entry *dst,
			   int nfheader_len)
{
	return 0;
}

static int xfrm4_fill_dst(struct xfrm_dst *xdst, struct net_device *dev)
{
	struct rtable *rt = (struct rtable *)xdst->route;

	xdst->u.rt.fl = rt->fl;

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

	xdst->u.rt.idev = in_dev_get(dev);
	if (!xdst->u.rt.idev)
		return -ENODEV;

	xdst->u.rt.peer = rt->peer;
	if (rt->peer)
		atomic_inc(&rt->peer->refcnt);

	/* Sheit... I remember I did this right. Apparently,
	 * it was magically lost, so this code needs audit */
	xdst->u.rt.rt_flags = rt->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST |
					      RTCF_LOCAL);
	xdst->u.rt.rt_type = rt->rt_type;
	xdst->u.rt.rt_src = rt->rt_src;
	xdst->u.rt.rt_dst = rt->rt_dst;
	xdst->u.rt.rt_gateway = rt->rt_gateway;
	xdst->u.rt.rt_spec_dst = rt->rt_spec_dst;

	return 0;
}

static void
_decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
{
	struct iphdr *iph = ip_hdr(skb);
	u8 *xprth = skb_network_header(skb) + iph->ihl * 4;

	memset(fl, 0, sizeof(struct flowi));
	if (!(iph->frag_off & htons(IP_MF | IP_OFFSET))) {
		switch (iph->protocol) {
		case IPPROTO_UDP:
		case IPPROTO_UDPLITE:
		case IPPROTO_TCP:
		case IPPROTO_SCTP:
		case IPPROTO_DCCP:
			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
				__be16 *ports = (__be16 *)xprth;

				fl->fl_ip_sport = ports[!!reverse];
				fl->fl_ip_dport = ports[!reverse];
			}
			break;

		case IPPROTO_ICMP:
			if (pskb_may_pull(skb, xprth + 2 - skb->data)) {
				u8 *icmp = xprth;

				fl->fl_icmp_type = icmp[0];
				fl->fl_icmp_code = icmp[1];
			}
			break;

		case IPPROTO_ESP:
			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
				__be32 *ehdr = (__be32 *)xprth;

				fl->fl_ipsec_spi = ehdr[0];
			}
			break;

		case IPPROTO_AH:
			if (pskb_may_pull(skb, xprth + 8 - skb->data)) {
				__be32 *ah_hdr = (__be32*)xprth;

				fl->fl_ipsec_spi = ah_hdr[1];
			}
			break;

		case IPPROTO_COMP:
			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
				__be16 *ipcomp_hdr = (__be16 *)xprth;

				fl->fl_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
			}
			break;
		default:
			fl->fl_ipsec_spi = 0;
			break;
		}
	}
	fl->proto = iph->protocol;
	fl->fl4_dst = reverse ? iph->saddr : iph->daddr;
	fl->fl4_src = reverse ? iph->daddr : iph->saddr;
	fl->fl4_tos = iph->tos;
}

static inline int xfrm4_garbage_collect(struct dst_ops *ops)
{
	xfrm4_policy_afinfo.garbage_collect();
	return (atomic_read(&xfrm4_dst_ops.entries) > xfrm4_dst_ops.gc_thresh*2);
}

static void xfrm4_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 xfrm4_dst_destroy(struct dst_entry *dst)
{
	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;

	if (likely(xdst->u.rt.idev))
		in_dev_put(xdst->u.rt.idev);
	if (likely(xdst->u.rt.peer))
		inet_putpeer(xdst->u.rt.peer);
	xfrm_dst_destroy(xdst);
}

static void xfrm4_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.rt.idev->dev == dev) {
		struct in_device *loopback_idev =
			in_dev_get(dev_net(dev)->loopback_dev);
		BUG_ON(!loopback_idev);

		do {
			in_dev_put(xdst->u.rt.idev);
			xdst->u.rt.idev = loopback_idev;
			in_dev_hold(loopback_idev);
			xdst = (struct xfrm_dst *)xdst->u.dst.child;
		} while (xdst->u.dst.xfrm);

		__in_dev_put(loopback_idev);
	}

	xfrm_dst_ifdown(dst, dev);
}

static struct dst_ops xfrm4_dst_ops = {
	.family =		AF_INET,
	.protocol =		__constant_htons(ETH_P_IP),
	.gc =			xfrm4_garbage_collect,
	.update_pmtu =		xfrm4_update_pmtu,
	.destroy =		xfrm4_dst_destroy,
	.ifdown =		xfrm4_dst_ifdown,
	.local_out =		__ip_local_out,
	.gc_thresh =		1024,
	.entry_size =		sizeof(struct xfrm_dst),
	.entries =		ATOMIC_INIT(0),
};

static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
	.family = 		AF_INET,
	.dst_ops =		&xfrm4_dst_ops,
	.dst_lookup =		xfrm4_dst_lookup,
	.get_saddr =		xfrm4_get_saddr,
	.find_bundle = 		__xfrm4_find_bundle,
	.decode_session =	_decode_session4,
	.get_tos =		xfrm4_get_tos,
	.init_path =		xfrm4_init_path,
	.fill_dst =		xfrm4_fill_dst,
};

static void __init xfrm4_policy_init(void)
{
	xfrm_policy_register_afinfo(&xfrm4_policy_afinfo);
}

static void __exit xfrm4_policy_fini(void)
{
	xfrm_policy_unregister_afinfo(&xfrm4_policy_afinfo);
}

void __init xfrm4_init(void)
{
	xfrm4_state_init();
	xfrm4_policy_init();
}
Commit	Line	Data
e905a9ed	1	/*
1da177e4 LT	2	* xfrm4_policy.c
	3	*
	4	* Changes:
	5	* Kazunori MIYAZAWA @USAGI
	6	* YOSHIFUJI Hideaki @USAGI
	7	* Split up af-specific portion
e905a9ed	8	*
1da177e4 LT	9	*/
1da177e4 LT	10
66cdb3ca HX	11	#include <linux/err.h>
66cdb3ca HX	12	#include <linux/kernel.h>
aabc9761	13	#include <linux/inetdevice.h>
45ff5a3f	14	#include <net/dst.h>
1da177e4 LT	15	#include <net/xfrm.h>
	16	#include <net/ip.h>
	17
	18	static struct dst_ops xfrm4_dst_ops;
	19	static struct xfrm_policy_afinfo xfrm4_policy_afinfo;
	20
66cdb3ca HX	21	static struct dst_entry xfrm4_dst_lookup(int tos, xfrm_address_t saddr,
66cdb3ca HX	22	xfrm_address_t *daddr)
1da177e4	23	{
66cdb3ca	24	struct flowi fl = {
a1e59abf PM	25	.nl_u = {
a1e59abf PM	26	.ip4_u = {
66cdb3ca	27	.tos = tos,
a1e59abf PM	28	.daddr = daddr->a4,
	29	},
	30	},
	31	};
66cdb3ca HX	32	struct dst_entry *dst;
	33	struct rtable *rt;
	34	int err;
a1e59abf	35
66cdb3ca HX	36	if (saddr)
	37	fl.fl4_src = saddr->a4;
	38
611c183e	39	err = __ip_route_output_key(&init_net, &rt, &fl);
66cdb3ca HX	40	dst = &rt->u.dst;
	41	if (err)
	42	dst = ERR_PTR(err);
	43	return dst;
	44	}
	45
	46	static int xfrm4_get_saddr(xfrm_address_t saddr, xfrm_address_t daddr)
	47	{
	48	struct dst_entry *dst;
	49	struct rtable *rt;
	50
	51	dst = xfrm4_dst_lookup(0, NULL, daddr);
	52	if (IS_ERR(dst))
	53	return -EHOSTUNREACH;
	54
	55	rt = (struct rtable *)dst;
	56	saddr->a4 = rt->rt_src;
	57	dst_release(dst);
	58	return 0;
a1e59abf PM	59	}
a1e59abf PM	60
1da177e4 LT	61	static struct dst_entry *
	62	__xfrm4_find_bundle(struct flowi fl, struct xfrm_policy policy)
	63	{
	64	struct dst_entry *dst;
	65
	66	read_lock_bh(&policy->lock);
	67	for (dst = policy->bundles; dst; dst = dst->next) {
	68	struct xfrm_dst xdst = (struct xfrm_dst)dst;
	69	if (xdst->u.rt.fl.oif == fl->oif && /XXX/
	70	xdst->u.rt.fl.fl4_dst == fl->fl4_dst &&
e905a9ed YH	71	xdst->u.rt.fl.fl4_src == fl->fl4_src &&
e905a9ed YH	72	xdst->u.rt.fl.fl4_tos == fl->fl4_tos &&
5b368e61	73	xfrm_bundle_ok(policy, xdst, fl, AF_INET, 0)) {
1da177e4 LT	74	dst_clone(dst);
	75	break;
	76	}
	77	}
	78	read_unlock_bh(&policy->lock);
	79	return dst;
	80	}
	81
25ee3286	82	static int xfrm4_get_tos(struct flowi *fl)
1da177e4	83	{
25ee3286 HX	84	return fl->fl4_tos;
25ee3286 HX	85	}
1da177e4	86
a1b05140 MN	87	static int xfrm4_init_path(struct xfrm_dst path, struct dst_entry dst,
	88	int nfheader_len)
	89	{
	90	return 0;
	91	}
	92
25ee3286 HX	93	static int xfrm4_fill_dst(struct xfrm_dst xdst, struct net_device dev)
	94	{
	95	struct rtable rt = (struct rtable )xdst->route;
1da177e4	96
25ee3286	97	xdst->u.rt.fl = rt->fl;
1da177e4	98
25ee3286 HX	99	xdst->u.dst.dev = dev;
25ee3286 HX	100	dev_hold(dev);
43372262	101
25ee3286 HX	102	xdst->u.rt.idev = in_dev_get(dev);
	103	if (!xdst->u.rt.idev)
	104	return -ENODEV;
1da177e4	105
25ee3286 HX	106	xdst->u.rt.peer = rt->peer;
	107	if (rt->peer)
	108	atomic_inc(&rt->peer->refcnt);
66cdb3ca	109
25ee3286 HX	110	/* Sheit... I remember I did this right. Apparently,
	111	* it was magically lost, so this code needs audit */
	112	xdst->u.rt.rt_flags = rt->rt_flags & (RTCF_BROADCAST \| RTCF_MULTICAST \|
	113	RTCF_LOCAL);
	114	xdst->u.rt.rt_type = rt->rt_type;
	115	xdst->u.rt.rt_src = rt->rt_src;
	116	xdst->u.rt.rt_dst = rt->rt_dst;
	117	xdst->u.rt.rt_gateway = rt->rt_gateway;
	118	xdst->u.rt.rt_spec_dst = rt->rt_spec_dst;
1da177e4	119
1da177e4	120	return 0;
1da177e4 LT	121	}
	122
	123	static void
d5422efe	124	_decode_session4(struct sk_buff skb, struct flowi fl, int reverse)
1da177e4	125	{
eddc9ec5	126	struct iphdr *iph = ip_hdr(skb);
d56f90a7	127	u8 xprth = skb_network_header(skb) + iph->ihl 4;
1da177e4 LT	128
	129	memset(fl, 0, sizeof(struct flowi));
	130	if (!(iph->frag_off & htons(IP_MF \| IP_OFFSET))) {
	131	switch (iph->protocol) {
	132	case IPPROTO_UDP:
ba4e58ec	133	case IPPROTO_UDPLITE:
1da177e4 LT	134	case IPPROTO_TCP:
1da177e4 LT	135	case IPPROTO_SCTP:
9e999993	136	case IPPROTO_DCCP:
1da177e4	137	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
8c689a6e	138	__be16 ports = (__be16 )xprth;
1da177e4	139
d5422efe HX	140	fl->fl_ip_sport = ports[!!reverse];
d5422efe HX	141	fl->fl_ip_dport = ports[!reverse];
1da177e4 LT	142	}
	143	break;
	144
	145	case IPPROTO_ICMP:
	146	if (pskb_may_pull(skb, xprth + 2 - skb->data)) {
	147	u8 *icmp = xprth;
	148
	149	fl->fl_icmp_type = icmp[0];
	150	fl->fl_icmp_code = icmp[1];
	151	}
	152	break;
	153
	154	case IPPROTO_ESP:
	155	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
4324a174	156	__be32 ehdr = (__be32 )xprth;
1da177e4 LT	157
	158	fl->fl_ipsec_spi = ehdr[0];
	159	}
	160	break;
	161
	162	case IPPROTO_AH:
	163	if (pskb_may_pull(skb, xprth + 8 - skb->data)) {
4324a174	164	__be32 ah_hdr = (__be32)xprth;
1da177e4 LT	165
	166	fl->fl_ipsec_spi = ah_hdr[1];
	167	}
	168	break;
	169
	170	case IPPROTO_COMP:
	171	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
4324a174	172	__be16 ipcomp_hdr = (__be16 )xprth;
1da177e4	173
4195f814	174	fl->fl_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
1da177e4 LT	175	}
	176	break;
	177	default:
	178	fl->fl_ipsec_spi = 0;
	179	break;
3ff50b79	180	}
1da177e4 LT	181	}
1da177e4 LT	182	fl->proto = iph->protocol;
d5422efe HX	183	fl->fl4_dst = reverse ? iph->saddr : iph->daddr;
d5422efe HX	184	fl->fl4_src = reverse ? iph->daddr : iph->saddr;
4da3089f	185	fl->fl4_tos = iph->tos;
1da177e4 LT	186	}
1da177e4 LT	187
569d3645	188	static inline int xfrm4_garbage_collect(struct dst_ops *ops)
1da177e4	189	{
1da177e4	190	xfrm4_policy_afinfo.garbage_collect();
1da177e4 LT	191	return (atomic_read(&xfrm4_dst_ops.entries) > xfrm4_dst_ops.gc_thresh*2);
	192	}
	193
	194	static void xfrm4_update_pmtu(struct dst_entry *dst, u32 mtu)
	195	{
	196	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	197	struct dst_entry *path = xdst->route;
	198
	199	path->ops->update_pmtu(path, mtu);
	200	}
	201
aabc9761 HX	202	static void xfrm4_dst_destroy(struct dst_entry *dst)
	203	{
	204	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	205
	206	if (likely(xdst->u.rt.idev))
	207	in_dev_put(xdst->u.rt.idev);
ed3e37dd	208	if (likely(xdst->u.rt.peer))
26db1677	209	inet_putpeer(xdst->u.rt.peer);
aabc9761 HX	210	xfrm_dst_destroy(xdst);
	211	}
	212
	213	static void xfrm4_dst_ifdown(struct dst_entry dst, struct net_device dev,
	214	int unregister)
	215	{
	216	struct xfrm_dst *xdst;
	217
	218	if (!unregister)
	219	return;
	220
	221	xdst = (struct xfrm_dst *)dst;
	222	if (xdst->u.rt.idev->dev == dev) {
5a3e55d6	223	struct in_device *loopback_idev =
c346dca1	224	in_dev_get(dev_net(dev)->loopback_dev);
aabc9761 HX	225	BUG_ON(!loopback_idev);
	226
	227	do {
	228	in_dev_put(xdst->u.rt.idev);
	229	xdst->u.rt.idev = loopback_idev;
	230	in_dev_hold(loopback_idev);
	231	xdst = (struct xfrm_dst *)xdst->u.dst.child;
	232	} while (xdst->u.dst.xfrm);
	233
	234	__in_dev_put(loopback_idev);
	235	}
	236
	237	xfrm_dst_ifdown(dst, dev);
	238	}
	239
1da177e4 LT	240	static struct dst_ops xfrm4_dst_ops = {
	241	.family = AF_INET,
	242	.protocol = __constant_htons(ETH_P_IP),
	243	.gc = xfrm4_garbage_collect,
	244	.update_pmtu = xfrm4_update_pmtu,
aabc9761 HX	245	.destroy = xfrm4_dst_destroy,
aabc9761 HX	246	.ifdown = xfrm4_dst_ifdown,
862b82c6	247	.local_out = __ip_local_out,
1da177e4 LT	248	.gc_thresh = 1024,
1da177e4 LT	249	.entry_size = sizeof(struct xfrm_dst),
e2422970	250	.entries = ATOMIC_INIT(0),
1da177e4 LT	251	};
	252
	253	static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
	254	.family = AF_INET,
1da177e4 LT	255	.dst_ops = &xfrm4_dst_ops,
1da177e4 LT	256	.dst_lookup = xfrm4_dst_lookup,
a1e59abf	257	.get_saddr = xfrm4_get_saddr,
1da177e4	258	.find_bundle = __xfrm4_find_bundle,
1da177e4	259	.decode_session = _decode_session4,
25ee3286	260	.get_tos = xfrm4_get_tos,
a1b05140	261	.init_path = xfrm4_init_path,
25ee3286	262	.fill_dst = xfrm4_fill_dst,
1da177e4 LT	263	};
	264
	265	static void __init xfrm4_policy_init(void)
	266	{
	267	xfrm_policy_register_afinfo(&xfrm4_policy_afinfo);
	268	}
	269
	270	static void __exit xfrm4_policy_fini(void)
	271	{
	272	xfrm_policy_unregister_afinfo(&xfrm4_policy_afinfo);
	273	}
	274
	275	void __init xfrm4_init(void)
	276	{
	277	xfrm4_state_init();
	278	xfrm4_policy_init();
	279	}
	280