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

static struct xfrm_policy_afinfo xfrm4_policy_afinfo;

static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4,
					    int tos, int oif,
					    const xfrm_address_t *saddr,
					    const xfrm_address_t *daddr)
{
	struct rtable *rt;

	memset(fl4, 0, sizeof(*fl4));
	fl4->daddr = daddr->a4;
	fl4->flowi4_tos = tos;
	fl4->flowi4_oif = oif;
	if (saddr)
		fl4->saddr = saddr->a4;

	rt = __ip_route_output_key(net, fl4);
	if (!IS_ERR(rt))
		return &rt->dst;

	return ERR_CAST(rt);
}

static struct dst_entry *xfrm4_dst_lookup(struct net *net, int tos, int oif,
					  const xfrm_address_t *saddr,
					  const xfrm_address_t *daddr)
{
	struct flowi4 fl4;

	return __xfrm4_dst_lookup(net, &fl4, tos, oif, saddr, daddr);
}

static int xfrm4_get_saddr(struct net *net, int oif,
			   xfrm_address_t *saddr, xfrm_address_t *daddr)
{
	struct dst_entry *dst;
	struct flowi4 fl4;

	dst = __xfrm4_dst_lookup(net, &fl4, 0, oif, NULL, daddr);
	if (IS_ERR(dst))
		return -EHOSTUNREACH;

	saddr->a4 = fl4.saddr;
	dst_release(dst);
	return 0;
}

static int xfrm4_get_tos(const struct flowi *fl)
{
	return IPTOS_RT_MASK & fl->u.ip4.flowi4_tos; /* Strip ECN bits */
}

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,
			  const struct flowi *fl)
{
	struct rtable *rt = (struct rtable *)xdst->route;
	const struct flowi4 *fl4 = &fl->u.ip4;

	xdst->u.rt.rt_iif = fl4->flowi4_iif;

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

	/* Sheit... I remember I did this right. Apparently,
	 * it was magically lost, so this code needs audit */
	xdst->u.rt.rt_is_input = rt->rt_is_input;
	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_gateway = rt->rt_gateway;
	xdst->u.rt.rt_uses_gateway = rt->rt_uses_gateway;
	xdst->u.rt.rt_pmtu = rt->rt_pmtu;
	INIT_LIST_HEAD(&xdst->u.rt.rt_uncached);

	return 0;
}

static void
_decode_session4(struct sk_buff *skb, struct flowi *fl, int reverse)
{
	const struct iphdr *iph = ip_hdr(skb);
	u8 *xprth = skb_network_header(skb) + iph->ihl * 4;
	struct flowi4 *fl4 = &fl->u.ip4;
	int oif = 0;

	if (skb_dst(skb)) {
		oif = vrf_master_ifindex(skb_dst(skb)->dev) ?
			: skb_dst(skb)->dev->ifindex;
	}

	memset(fl4, 0, sizeof(struct flowi4));
	fl4->flowi4_mark = skb->mark;
	fl4->flowi4_oif = reverse ? skb->skb_iif : oif;

	if (!ip_is_fragment(iph)) {
		switch (iph->protocol) {
		case IPPROTO_UDP:
		case IPPROTO_UDPLITE:
		case IPPROTO_TCP:
		case IPPROTO_SCTP:
		case IPPROTO_DCCP:
			if (xprth + 4 < skb->data ||
			    pskb_may_pull(skb, xprth + 4 - skb->data)) {
				__be16 *ports = (__be16 *)xprth;

				fl4->fl4_sport = ports[!!reverse];
				fl4->fl4_dport = ports[!reverse];
			}
			break;

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

				fl4->fl4_icmp_type = icmp[0];
				fl4->fl4_icmp_code = icmp[1];
			}
			break;

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

				fl4->fl4_ipsec_spi = ehdr[0];
			}
			break;

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

				fl4->fl4_ipsec_spi = ah_hdr[1];
			}
			break;

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

				fl4->fl4_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
			}
			break;

		case IPPROTO_GRE:
			if (pskb_may_pull(skb, xprth + 12 - skb->data)) {
				__be16 *greflags = (__be16 *)xprth;
				__be32 *gre_hdr = (__be32 *)xprth;

				if (greflags[0] & GRE_KEY) {
					if (greflags[0] & GRE_CSUM)
						gre_hdr++;
					fl4->fl4_gre_key = gre_hdr[1];
				}
			}
			break;

		default:
			fl4->fl4_ipsec_spi = 0;
			break;
		}
	}
	fl4->flowi4_proto = iph->protocol;
	fl4->daddr = reverse ? iph->saddr : iph->daddr;
	fl4->saddr = reverse ? iph->daddr : iph->saddr;
	fl4->flowi4_tos = iph->tos;
}

static inline int xfrm4_garbage_collect(struct dst_ops *ops)
{
	struct net *net = container_of(ops, struct net, xfrm.xfrm4_dst_ops);

	xfrm4_policy_afinfo.garbage_collect(net);
	return (dst_entries_get_slow(ops) > ops->gc_thresh * 2);
}

static void xfrm4_update_pmtu(struct dst_entry *dst, struct sock *sk,
			      struct sk_buff *skb, u32 mtu)
{
	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
	struct dst_entry *path = xdst->route;

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

static void xfrm4_redirect(struct dst_entry *dst, struct sock *sk,
			   struct sk_buff *skb)
{
	struct xfrm_dst *xdst = (struct xfrm_dst *)dst;
	struct dst_entry *path = xdst->route;

	path->ops->redirect(path, sk, skb);
}

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

	dst_destroy_metrics_generic(dst);

	xfrm_dst_destroy(xdst);
}

static void xfrm4_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
			     int unregister)
{
	if (!unregister)
		return;

	xfrm_dst_ifdown(dst, dev);
}

static struct dst_ops xfrm4_dst_ops = {
	.family =		AF_INET,
	.gc =			xfrm4_garbage_collect,
	.update_pmtu =		xfrm4_update_pmtu,
	.redirect =		xfrm4_redirect,
	.cow_metrics =		dst_cow_metrics_generic,
	.destroy =		xfrm4_dst_destroy,
	.ifdown =		xfrm4_dst_ifdown,
	.local_out =		__ip_local_out,
	.gc_thresh =		32768,
};

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,
	.decode_session =	_decode_session4,
	.get_tos =		xfrm4_get_tos,
	.init_path =		xfrm4_init_path,
	.fill_dst =		xfrm4_fill_dst,
	.blackhole_route =	ipv4_blackhole_route,
};

#ifdef CONFIG_SYSCTL
static struct ctl_table xfrm4_policy_table[] = {
	{
		.procname       = "xfrm4_gc_thresh",
		.data           = &init_net.xfrm.xfrm4_dst_ops.gc_thresh,
		.maxlen         = sizeof(int),
		.mode           = 0644,
		.proc_handler   = proc_dointvec,
	},
	{ }
};

static int __net_init xfrm4_net_init(struct net *net)
{
	struct ctl_table *table;
	struct ctl_table_header *hdr;

	table = xfrm4_policy_table;
	if (!net_eq(net, &init_net)) {
		table = kmemdup(table, sizeof(xfrm4_policy_table), GFP_KERNEL);
		if (!table)
			goto err_alloc;

		table[0].data = &net->xfrm.xfrm4_dst_ops.gc_thresh;
	}

	hdr = register_net_sysctl(net, "net/ipv4", table);
	if (!hdr)
		goto err_reg;

	net->ipv4.xfrm4_hdr = hdr;
	return 0;

err_reg:
	if (!net_eq(net, &init_net))
		kfree(table);
err_alloc:
	return -ENOMEM;
}

static void __net_exit xfrm4_net_exit(struct net *net)
{
	struct ctl_table *table;

	if (!net->ipv4.xfrm4_hdr)
		return;

	table = net->ipv4.xfrm4_hdr->ctl_table_arg;
	unregister_net_sysctl_table(net->ipv4.xfrm4_hdr);
	if (!net_eq(net, &init_net))
		kfree(table);
}

static struct pernet_operations __net_initdata xfrm4_net_ops = {
	.init	= xfrm4_net_init,
	.exit	= xfrm4_net_exit,
};
#endif

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

void __init xfrm4_init(void)
{
	dst_entries_init(&xfrm4_dst_ops);

	xfrm4_state_init();
	xfrm4_policy_init();
	xfrm4_protocol_init();
#ifdef CONFIG_SYSCTL
	register_pernet_subsys(&xfrm4_net_ops);
#endif
}
Commit	Line	Data
	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
	11	#include <linux/err.h>
	12	#include <linux/kernel.h>
	13	#include <linux/inetdevice.h>
	14	#include <linux/if_tunnel.h>
	15	#include <net/dst.h>
	16	#include <net/xfrm.h>
	17	#include <net/ip.h>
	18	#include <net/vrf.h>
	19
	20	static struct xfrm_policy_afinfo xfrm4_policy_afinfo;
	21
	22	static struct dst_entry __xfrm4_dst_lookup(struct net net, struct flowi4 *fl4,
	23	int tos, int oif,
	24	const xfrm_address_t *saddr,
	25	const xfrm_address_t *daddr)
	26	{
	27	struct rtable *rt;
	28
	29	memset(fl4, 0, sizeof(*fl4));
	30	fl4->daddr = daddr->a4;
	31	fl4->flowi4_tos = tos;
	32	fl4->flowi4_oif = oif;
	33	if (saddr)
	34	fl4->saddr = saddr->a4;
	35
	36	rt = __ip_route_output_key(net, fl4);
	37	if (!IS_ERR(rt))
	38	return &rt->dst;
	39
	40	return ERR_CAST(rt);
	41	}
	42
	43	static struct dst_entry xfrm4_dst_lookup(struct net net, int tos, int oif,
	44	const xfrm_address_t *saddr,
	45	const xfrm_address_t *daddr)
	46	{
	47	struct flowi4 fl4;
	48
	49	return __xfrm4_dst_lookup(net, &fl4, tos, oif, saddr, daddr);
	50	}
	51
	52	static int xfrm4_get_saddr(struct net *net, int oif,
	53	xfrm_address_t saddr, xfrm_address_t daddr)
	54	{
	55	struct dst_entry *dst;
	56	struct flowi4 fl4;
	57
	58	dst = __xfrm4_dst_lookup(net, &fl4, 0, oif, NULL, daddr);
	59	if (IS_ERR(dst))
	60	return -EHOSTUNREACH;
	61
	62	saddr->a4 = fl4.saddr;
	63	dst_release(dst);
	64	return 0;
	65	}
	66
	67	static int xfrm4_get_tos(const struct flowi *fl)
	68	{
	69	return IPTOS_RT_MASK & fl->u.ip4.flowi4_tos; /* Strip ECN bits */
	70	}
	71
	72	static int xfrm4_init_path(struct xfrm_dst path, struct dst_entry dst,
	73	int nfheader_len)
	74	{
	75	return 0;
	76	}
	77
	78	static int xfrm4_fill_dst(struct xfrm_dst xdst, struct net_device dev,
	79	const struct flowi *fl)
	80	{
	81	struct rtable rt = (struct rtable )xdst->route;
	82	const struct flowi4 *fl4 = &fl->u.ip4;
	83
	84	xdst->u.rt.rt_iif = fl4->flowi4_iif;
	85
	86	xdst->u.dst.dev = dev;
	87	dev_hold(dev);
	88
	89	/* Sheit... I remember I did this right. Apparently,
	90	* it was magically lost, so this code needs audit */
	91	xdst->u.rt.rt_is_input = rt->rt_is_input;
	92	xdst->u.rt.rt_flags = rt->rt_flags & (RTCF_BROADCAST \| RTCF_MULTICAST \|
	93	RTCF_LOCAL);
	94	xdst->u.rt.rt_type = rt->rt_type;
	95	xdst->u.rt.rt_gateway = rt->rt_gateway;
	96	xdst->u.rt.rt_uses_gateway = rt->rt_uses_gateway;
	97	xdst->u.rt.rt_pmtu = rt->rt_pmtu;
	98	INIT_LIST_HEAD(&xdst->u.rt.rt_uncached);
	99
	100	return 0;
	101	}
	102
	103	static void
	104	_decode_session4(struct sk_buff skb, struct flowi fl, int reverse)
	105	{
	106	const struct iphdr *iph = ip_hdr(skb);
	107	u8 xprth = skb_network_header(skb) + iph->ihl 4;
	108	struct flowi4 *fl4 = &fl->u.ip4;
	109	int oif = 0;
	110
	111	if (skb_dst(skb)) {
	112	oif = vrf_master_ifindex(skb_dst(skb)->dev) ?
	113	: skb_dst(skb)->dev->ifindex;
	114	}
	115
	116	memset(fl4, 0, sizeof(struct flowi4));
	117	fl4->flowi4_mark = skb->mark;
	118	fl4->flowi4_oif = reverse ? skb->skb_iif : oif;
	119
	120	if (!ip_is_fragment(iph)) {
	121	switch (iph->protocol) {
	122	case IPPROTO_UDP:
	123	case IPPROTO_UDPLITE:
	124	case IPPROTO_TCP:
	125	case IPPROTO_SCTP:
	126	case IPPROTO_DCCP:
	127	if (xprth + 4 < skb->data \|\|
	128	pskb_may_pull(skb, xprth + 4 - skb->data)) {
	129	__be16 ports = (__be16 )xprth;
	130
	131	fl4->fl4_sport = ports[!!reverse];
	132	fl4->fl4_dport = ports[!reverse];
	133	}
	134	break;
	135
	136	case IPPROTO_ICMP:
	137	if (pskb_may_pull(skb, xprth + 2 - skb->data)) {
	138	u8 *icmp = xprth;
	139
	140	fl4->fl4_icmp_type = icmp[0];
	141	fl4->fl4_icmp_code = icmp[1];
	142	}
	143	break;
	144
	145	case IPPROTO_ESP:
	146	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
	147	__be32 ehdr = (__be32 )xprth;
	148
	149	fl4->fl4_ipsec_spi = ehdr[0];
	150	}
	151	break;
	152
	153	case IPPROTO_AH:
	154	if (pskb_may_pull(skb, xprth + 8 - skb->data)) {
	155	__be32 ah_hdr = (__be32 )xprth;
	156
	157	fl4->fl4_ipsec_spi = ah_hdr[1];
	158	}
	159	break;
	160
	161	case IPPROTO_COMP:
	162	if (pskb_may_pull(skb, xprth + 4 - skb->data)) {
	163	__be16 ipcomp_hdr = (__be16 )xprth;
	164
	165	fl4->fl4_ipsec_spi = htonl(ntohs(ipcomp_hdr[1]));
	166	}
	167	break;
	168
	169	case IPPROTO_GRE:
	170	if (pskb_may_pull(skb, xprth + 12 - skb->data)) {
	171	__be16 greflags = (__be16 )xprth;
	172	__be32 gre_hdr = (__be32 )xprth;
	173
	174	if (greflags[0] & GRE_KEY) {
	175	if (greflags[0] & GRE_CSUM)
	176	gre_hdr++;
	177	fl4->fl4_gre_key = gre_hdr[1];
	178	}
	179	}
	180	break;
	181
	182	default:
	183	fl4->fl4_ipsec_spi = 0;
	184	break;
	185	}
	186	}
	187	fl4->flowi4_proto = iph->protocol;
	188	fl4->daddr = reverse ? iph->saddr : iph->daddr;
	189	fl4->saddr = reverse ? iph->daddr : iph->saddr;
	190	fl4->flowi4_tos = iph->tos;
	191	}
	192
	193	static inline int xfrm4_garbage_collect(struct dst_ops *ops)
	194	{
	195	struct net *net = container_of(ops, struct net, xfrm.xfrm4_dst_ops);
	196
	197	xfrm4_policy_afinfo.garbage_collect(net);
	198	return (dst_entries_get_slow(ops) > ops->gc_thresh * 2);
	199	}
	200
	201	static void xfrm4_update_pmtu(struct dst_entry dst, struct sock sk,
	202	struct sk_buff *skb, u32 mtu)
	203	{
	204	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	205	struct dst_entry *path = xdst->route;
	206
	207	path->ops->update_pmtu(path, sk, skb, mtu);
	208	}
	209
	210	static void xfrm4_redirect(struct dst_entry dst, struct sock sk,
	211	struct sk_buff *skb)
	212	{
	213	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	214	struct dst_entry *path = xdst->route;
	215
	216	path->ops->redirect(path, sk, skb);
	217	}
	218
	219	static void xfrm4_dst_destroy(struct dst_entry *dst)
	220	{
	221	struct xfrm_dst xdst = (struct xfrm_dst )dst;
	222
	223	dst_destroy_metrics_generic(dst);
	224
	225	xfrm_dst_destroy(xdst);
	226	}
	227
	228	static void xfrm4_dst_ifdown(struct dst_entry dst, struct net_device dev,
	229	int unregister)
	230	{
	231	if (!unregister)
	232	return;
	233
	234	xfrm_dst_ifdown(dst, dev);
	235	}
	236
	237	static struct dst_ops xfrm4_dst_ops = {
	238	.family = AF_INET,
	239	.gc = xfrm4_garbage_collect,
	240	.update_pmtu = xfrm4_update_pmtu,
	241	.redirect = xfrm4_redirect,
	242	.cow_metrics = dst_cow_metrics_generic,
	243	.destroy = xfrm4_dst_destroy,
	244	.ifdown = xfrm4_dst_ifdown,
	245	.local_out = __ip_local_out,
	246	.gc_thresh = 32768,
	247	};
	248
	249	static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
	250	.family = AF_INET,
	251	.dst_ops = &xfrm4_dst_ops,
	252	.dst_lookup = xfrm4_dst_lookup,
	253	.get_saddr = xfrm4_get_saddr,
	254	.decode_session = _decode_session4,
	255	.get_tos = xfrm4_get_tos,
	256	.init_path = xfrm4_init_path,
	257	.fill_dst = xfrm4_fill_dst,
	258	.blackhole_route = ipv4_blackhole_route,
	259	};
	260
	261	#ifdef CONFIG_SYSCTL
	262	static struct ctl_table xfrm4_policy_table[] = {
	263	{
	264	.procname = "xfrm4_gc_thresh",
	265	.data = &init_net.xfrm.xfrm4_dst_ops.gc_thresh,
	266	.maxlen = sizeof(int),
	267	.mode = 0644,
	268	.proc_handler = proc_dointvec,
	269	},
	270	{ }
	271	};
	272
	273	static int __net_init xfrm4_net_init(struct net *net)
	274	{
	275	struct ctl_table *table;
	276	struct ctl_table_header *hdr;
	277
	278	table = xfrm4_policy_table;
	279	if (!net_eq(net, &init_net)) {
	280	table = kmemdup(table, sizeof(xfrm4_policy_table), GFP_KERNEL);
	281	if (!table)
	282	goto err_alloc;
	283
	284	table[0].data = &net->xfrm.xfrm4_dst_ops.gc_thresh;
	285	}
	286
	287	hdr = register_net_sysctl(net, "net/ipv4", table);
	288	if (!hdr)
	289	goto err_reg;
	290
	291	net->ipv4.xfrm4_hdr = hdr;
	292	return 0;
	293
	294	err_reg:
	295	if (!net_eq(net, &init_net))
	296	kfree(table);
	297	err_alloc:
	298	return -ENOMEM;
	299	}
	300
	301	static void __net_exit xfrm4_net_exit(struct net *net)
	302	{
	303	struct ctl_table *table;
	304
	305	if (!net->ipv4.xfrm4_hdr)
	306	return;
	307
	308	table = net->ipv4.xfrm4_hdr->ctl_table_arg;
	309	unregister_net_sysctl_table(net->ipv4.xfrm4_hdr);
	310	if (!net_eq(net, &init_net))
	311	kfree(table);
	312	}
	313
	314	static struct pernet_operations __net_initdata xfrm4_net_ops = {
	315	.init = xfrm4_net_init,
	316	.exit = xfrm4_net_exit,
	317	};
	318	#endif
	319
	320	static void __init xfrm4_policy_init(void)
	321	{
	322	xfrm_policy_register_afinfo(&xfrm4_policy_afinfo);
	323	}
	324
	325	void __init xfrm4_init(void)
	326	{
	327	dst_entries_init(&xfrm4_dst_ops);
	328
	329	xfrm4_state_init();
	330	xfrm4_policy_init();
	331	xfrm4_protocol_init();
	332	#ifdef CONFIG_SYSCTL
	333	register_pernet_subsys(&xfrm4_net_ops);
	334	#endif
	335	}
	336