[deliverable/linux.git] / net / ipv4 / xfrm4_policy.c

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

#include <linux/compiler.h>
#include <linux/config.h>
#include <linux/inetdevice.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 xfrm_type_map xfrm4_type_map = { .lock = RW_LOCK_UNLOCKED };

static int xfrm4_dst_lookup(struct xfrm_dst **dst, struct flowi *fl)
{
	return __ip_route_output_key((struct rtable**)dst, fl);
}

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 &&
		    xfrm_bundle_ok(xdst, fl, AF_INET)) {
			dst_clone(dst);
			break;
		}
	}
	read_unlock_bh(&policy->lock);
	return dst;
}

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

static int
__xfrm4_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 rtable *rt0 = (struct rtable*)(*dst_p);
	struct rtable *rt = rt0;
	u32 remote = fl->fl4_dst;
	u32 local  = fl->fl4_src;
	struct flowi fl_tunnel = {
		.nl_u = {
			.ip4_u = {
				.saddr = local,
				.daddr = remote
			}
		}
	};
	int i;
	int err;
	int header_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(&xfrm4_dst_ops);
		struct xfrm_dst *xdst;
		int tunnel = 0;

		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;

		dst1->next = dst_prev;
		dst_prev = dst1;
		if (xfrm[i]->props.mode) {
			remote = xfrm[i]->id.daddr.a4;
			local  = xfrm[i]->props.saddr.a4;
			tunnel = 1;
		}
		header_len += xfrm[i]->props.header_len;
		trailer_len += xfrm[i]->props.trailer_len;

		if (tunnel) {
			fl_tunnel.fl4_src = local;
			fl_tunnel.fl4_dst = remote;
			err = xfrm_dst_lookup((struct xfrm_dst **)&rt,
					      &fl_tunnel, AF_INET);
			if (err)
				goto error;
		} else
			dst_hold(&rt->u.dst);
	}

	dst_prev->child = &rt->u.dst;
	dst->path = &rt->u.dst;

	*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;
		x->u.rt.fl = *fl;

		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->trailer_len	= trailer_len;
		memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));

		/* Copy neighbout for reachability confirmation */
		dst_prev->neighbour	= neigh_clone(rt->u.dst.neighbour);
		dst_prev->input		= rt->u.dst.input;
		dst_prev->output	= xfrm4_output;
		if (rt->peer)
			atomic_inc(&rt->peer->refcnt);
		x->u.rt.peer = rt->peer;
		/* Sheit... I remember I did this right. Apparently,
		 * it was magically lost, so this code needs audit */
		x->u.rt.rt_flags = rt0->rt_flags&(RTCF_BROADCAST|RTCF_MULTICAST|RTCF_LOCAL);
		x->u.rt.rt_type = rt->rt_type;
		x->u.rt.rt_src = rt0->rt_src;
		x->u.rt.rt_dst = rt0->rt_dst;
		x->u.rt.rt_gateway = rt->rt_gateway;
		x->u.rt.rt_spec_dst = rt0->rt_spec_dst;
		x->u.rt.idev = rt0->idev;
		in_dev_hold(rt0->idev);
		header_len -= x->u.dst.xfrm->props.header_len;
		trailer_len -= x->u.dst.xfrm->props.trailer_len;
	}

	xfrm_init_pmtu(dst);
	return 0;

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

static void
_decode_session4(struct sk_buff *skb, struct flowi *fl)
{
	struct iphdr *iph = skb->nh.iph;
	u8 *xprth = skb->nh.raw + 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_TCP:
		case IPPROTO_SCTP:
		case IPPROTO_DCCP:
			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
				u16 *ports = (u16 *)xprth;

				fl->fl_ip_sport = ports[0];
				fl->fl_ip_dport = ports[1];
			}
			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)) {
				u32 *ehdr = (u32 *)xprth;

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

		case IPPROTO_AH:
			if (pskb_may_pull(skb, xprth + 8 - skb->data)) {
				u32 *ah_hdr = (u32*)xprth;

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

		case IPPROTO_COMP:
			if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
				u16 *ipcomp_hdr = (u16 *)xprth;

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

static inline int xfrm4_garbage_collect(void)
{
	read_lock(&xfrm4_policy_afinfo.lock);
	xfrm4_policy_afinfo.garbage_collect();
	read_unlock(&xfrm4_policy_afinfo.lock);
	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);
	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(&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,
	.gc_thresh =		1024,
	.entry_size =		sizeof(struct xfrm_dst),
};

static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
	.family = 		AF_INET,
	.lock = 		RW_LOCK_UNLOCKED,
	.type_map = 		&xfrm4_type_map,
	.dst_ops =		&xfrm4_dst_ops,
	.dst_lookup =		xfrm4_dst_lookup,
	.find_bundle = 		__xfrm4_find_bundle,
	.bundle_create =	__xfrm4_bundle_create,
	.decode_session =	_decode_session4,
};

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
1da177e4 LT	1	/*
	2	* xfrm4_policy.c
	3	*
	4	* Changes:
	5	* Kazunori MIYAZAWA @USAGI
	6	* YOSHIFUJI Hideaki @USAGI
	7	* Split up af-specific portion
	8	*
	9	*/
	10
aabc9761	11	#include <linux/compiler.h>
1da177e4	12	#include <linux/config.h>
aabc9761	13	#include <linux/inetdevice.h>
1da177e4 LT	14	#include <net/xfrm.h>
	15	#include <net/ip.h>
	16
	17	static struct dst_ops xfrm4_dst_ops;
	18	static struct xfrm_policy_afinfo xfrm4_policy_afinfo;
	19
	20	static struct xfrm_type_map xfrm4_type_map = { .lock = RW_LOCK_UNLOCKED };
	21
	22	static int xfrm4_dst_lookup(struct xfrm_dst *dst, struct flowi fl)
	23	{
	24	return __ip_route_output_key((struct rtable**)dst, fl);
	25	}
	26
	27	static struct dst_entry *
	28	__xfrm4_find_bundle(struct flowi fl, struct xfrm_policy policy)
	29	{
	30	struct dst_entry *dst;
	31
	32	read_lock_bh(&policy->lock);
	33	for (dst = policy->bundles; dst; dst = dst->next) {
	34	struct xfrm_dst xdst = (struct xfrm_dst)dst;
	35	if (xdst->u.rt.fl.oif == fl->oif && /XXX/
	36	xdst->u.rt.fl.fl4_dst == fl->fl4_dst &&
	37	xdst->u.rt.fl.fl4_src == fl->fl4_src &&
	38	xfrm_bundle_ok(xdst, fl, AF_INET)) {
	39	dst_clone(dst);
	40	break;
	41	}
	42	}
	43	read_unlock_bh(&policy->lock);
	44	return dst;
	45	}
	46
	47	/* Allocate chain of dst_entry's, attach known xfrm's, calculate
	48	* all the metrics... Shortly, bundle a bundle.
	49	*/
	50
	51	static int
	52	__xfrm4_bundle_create(struct xfrm_policy policy, struct xfrm_state *xfrm, int nx,
	53	struct flowi fl, struct dst_entry *dst_p)
	54	{
	55	struct dst_entry dst, dst_prev;
	56	struct rtable rt0 = (struct rtable)(*dst_p);
	57	struct rtable *rt = rt0;
	58	u32 remote = fl->fl4_dst;
	59	u32 local = fl->fl4_src;
	60	struct flowi fl_tunnel = {
	61	.nl_u = {
	62	.ip4_u = {
	63	.saddr = local,
	64	.daddr = remote
	65	}
	66	}
	67	};
	68	int i;
	69	int err;
	70	int header_len = 0;
	71	int trailer_len = 0;
	72
	73	dst = dst_prev = NULL;
	74	dst_hold(&rt->u.dst);
	75
	76	for (i = 0; i < nx; i++) {
	77	struct dst_entry *dst1 = dst_alloc(&xfrm4_dst_ops);
78	struct xfrm_dst *xdst;
79	int tunnel = 0;
80
81	if (unlikely(dst1 == NULL)) {
82	err = -ENOBUFS;
83	dst_release(&rt->u.dst);
84	goto error;
85	}
86
87	if (!dst)
88	dst = dst1;
89	else {
90	dst_prev->child = dst1;
91	dst1->flags \|= DST_NOHASH;
92	dst_clone(dst1);
93	}
94
95	xdst = (struct xfrm_dst *)dst1;
96	xdst->route = &rt->u.dst;
97
98	dst1->next = dst_prev;
99	dst_prev = dst1;
100	if (xfrm[i]->props.mode) {
101	remote = xfrm[i]->id.daddr.a4;
102	local = xfrm[i]->props.saddr.a4;
103	tunnel = 1;
104	}
105	header_len += xfrm[i]->props.header_len;
106	trailer_len += xfrm[i]->props.trailer_len;
107
108	if (tunnel) {
109	fl_tunnel.fl4_src = local;
110	fl_tunnel.fl4_dst = remote;
111	err = xfrm_dst_lookup((struct xfrm_dst **)&rt,
112	&fl_tunnel, AF_INET);
113	if (err)
114	goto error;
115	} else
116	dst_hold(&rt->u.dst);
117	}
118
119	dst_prev->child = &rt->u.dst;
120	dst->path = &rt->u.dst;
121
122	*dst_p = dst;
123	dst = dst_prev;
124
125	dst_prev = *dst_p;
126	i = 0;
127	for (; dst_prev != &rt->u.dst; dst_prev = dst_prev->child) {
128	struct xfrm_dst x = (struct xfrm_dst)dst_prev;
129	x->u.rt.fl = *fl;
130
131	dst_prev->xfrm = xfrm[i++];
132	dst_prev->dev = rt->u.dst.dev;
133	if (rt->u.dst.dev)
134	dev_hold(rt->u.dst.dev);
135	dst_prev->obsolete = -1;
136	dst_prev->flags \|= DST_HOST;
137	dst_prev->lastuse = jiffies;
138	dst_prev->header_len = header_len;
139	dst_prev->trailer_len = trailer_len;
140	memcpy(&dst_prev->metrics, &x->route->metrics, sizeof(dst_prev->metrics));
141
142	/* Copy neighbout for reachability confirmation */
143	dst_prev->neighbour = neigh_clone(rt->u.dst.neighbour);
144	dst_prev->input = rt->u.dst.input;
145	dst_prev->output = xfrm4_output;
146	if (rt->peer)
147	atomic_inc(&rt->peer->refcnt);
148	x->u.rt.peer = rt->peer;
149	/* Sheit... I remember I did this right. Apparently,
150	* it was magically lost, so this code needs audit */
151	x->u.rt.rt_flags = rt0->rt_flags&(RTCF_BROADCAST\|RTCF_MULTICAST\|RTCF_LOCAL);
152	x->u.rt.rt_type = rt->rt_type;
153	x->u.rt.rt_src = rt0->rt_src;
154	x->u.rt.rt_dst = rt0->rt_dst;
155	x->u.rt.rt_gateway = rt->rt_gateway;
156	x->u.rt.rt_spec_dst = rt0->rt_spec_dst;
aabc9761 HX	157	x->u.rt.idev = rt0->idev;
aabc9761 HX	158	in_dev_hold(rt0->idev);
1da177e4 LT	159	header_len -= x->u.dst.xfrm->props.header_len;
	160	trailer_len -= x->u.dst.xfrm->props.trailer_len;
	161	}
	162
	163	xfrm_init_pmtu(dst);
	164	return 0;
	165
	166	error:
	167	if (dst)
	168	dst_free(dst);
	169	return err;
	170	}
	171
	172	static void
	173	_decode_session4(struct sk_buff skb, struct flowi fl)
	174	{
	175	struct iphdr *iph = skb->nh.iph;
	176	u8 xprth = skb->nh.raw + iph->ihl4;
	177
	178	memset(fl, 0, sizeof(struct flowi));
	179	if (!(iph->frag_off & htons(IP_MF \| IP_OFFSET))) {
	180	switch (iph->protocol) {
	181	case IPPROTO_UDP:
	182	case IPPROTO_TCP:
	183	case IPPROTO_SCTP:
9e999993	184	case IPPROTO_DCCP:
1da177e4 LT	185	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
	186	u16 ports = (u16 )xprth;
	187
	188	fl->fl_ip_sport = ports[0];
	189	fl->fl_ip_dport = ports[1];
	190	}
	191	break;
	192
	193	case IPPROTO_ICMP:
	194	if (pskb_may_pull(skb, xprth + 2 - skb->data)) {
	195	u8 *icmp = xprth;
	196
	197	fl->fl_icmp_type = icmp[0];
	198	fl->fl_icmp_code = icmp[1];
	199	}
	200	break;
	201
	202	case IPPROTO_ESP:
	203	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
	204	u32 ehdr = (u32 )xprth;
	205
	206	fl->fl_ipsec_spi = ehdr[0];
	207	}
	208	break;
	209
	210	case IPPROTO_AH:
	211	if (pskb_may_pull(skb, xprth + 8 - skb->data)) {
	212	u32 ah_hdr = (u32)xprth;
	213
	214	fl->fl_ipsec_spi = ah_hdr[1];
	215	}
	216	break;
	217
	218	case IPPROTO_COMP:
	219	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
	220	u16 ipcomp_hdr = (u16 )xprth;
	221
	222	fl->fl_ipsec_spi = ntohl(ntohs(ipcomp_hdr[1]));
	223	}
	224	break;
	225	default:
	226	fl->fl_ipsec_spi = 0;
	227	break;
	228	};
	229	}
	230	fl->proto = iph->protocol;
	231	fl->fl4_dst = iph->daddr;
	232	fl->fl4_src = iph->saddr;
	233	}
	234
	235	static inline int xfrm4_garbage_collect(void)
	236	{
	237	read_lock(&xfrm4_policy_afinfo.lock);
	238	xfrm4_policy_afinfo.garbage_collect();
	239	read_unlock(&xfrm4_policy_afinfo.lock);
	240	return (atomic_read(&xfrm4_dst_ops.entries) > xfrm4_dst_ops.gc_thresh*2);
	241	}
	242
	243	static void xfrm4_update_pmtu(struct dst_entry *dst, u32 mtu)
	244	{
	245	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	246	struct dst_entry *path = xdst->route;
	247
	248	path->ops->update_pmtu(path, mtu);
249	}
250
aabc9761 HX	251	static void xfrm4_dst_destroy(struct dst_entry *dst)
	252	{
	253	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	254
	255	if (likely(xdst->u.rt.idev))
	256	in_dev_put(xdst->u.rt.idev);
	257	xfrm_dst_destroy(xdst);
	258	}
	259
	260	static void xfrm4_dst_ifdown(struct dst_entry dst, struct net_device dev,
	261	int unregister)
	262	{
	263	struct xfrm_dst *xdst;
	264
	265	if (!unregister)
	266	return;
	267
	268	xdst = (struct xfrm_dst *)dst;
	269	if (xdst->u.rt.idev->dev == dev) {
	270	struct in_device *loopback_idev = in_dev_get(&loopback_dev);
	271	BUG_ON(!loopback_idev);
	272
	273	do {
	274	in_dev_put(xdst->u.rt.idev);
	275	xdst->u.rt.idev = loopback_idev;
	276	in_dev_hold(loopback_idev);
	277	xdst = (struct xfrm_dst *)xdst->u.dst.child;
	278	} while (xdst->u.dst.xfrm);
	279
	280	__in_dev_put(loopback_idev);
	281	}
	282
	283	xfrm_dst_ifdown(dst, dev);
	284	}
	285
1da177e4 LT	286	static struct dst_ops xfrm4_dst_ops = {
	287	.family = AF_INET,
	288	.protocol = __constant_htons(ETH_P_IP),
	289	.gc = xfrm4_garbage_collect,
	290	.update_pmtu = xfrm4_update_pmtu,
aabc9761 HX	291	.destroy = xfrm4_dst_destroy,
aabc9761 HX	292	.ifdown = xfrm4_dst_ifdown,
1da177e4 LT	293	.gc_thresh = 1024,
	294	.entry_size = sizeof(struct xfrm_dst),
	295	};
	296
	297	static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
	298	.family = AF_INET,
	299	.lock = RW_LOCK_UNLOCKED,
	300	.type_map = &xfrm4_type_map,
	301	.dst_ops = &xfrm4_dst_ops,
	302	.dst_lookup = xfrm4_dst_lookup,
	303	.find_bundle = __xfrm4_find_bundle,
	304	.bundle_create = __xfrm4_bundle_create,
	305	.decode_session = _decode_session4,
	306	};
	307
	308	static void __init xfrm4_policy_init(void)
	309	{
	310	xfrm_policy_register_afinfo(&xfrm4_policy_afinfo);
	311	}
	312
	313	static void __exit xfrm4_policy_fini(void)
	314	{
	315	xfrm_policy_unregister_afinfo(&xfrm4_policy_afinfo);
	316	}
	317
	318	void __init xfrm4_init(void)
	319	{
	320	xfrm4_state_init();
	321	xfrm4_policy_init();
	322	}
	323