[IPV6]: ROUTE: Add experimental support for Route Information Option in RA (RFC4191).

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/ipv6_route.h

@@ -23,6 +23,8 @@
 #define RTF_NONEXTHOP	0x00200000	/* route with no nexthop	*/
 #define RTF_EXPIRES	0x00400000
 
+#define RTF_ROUTEINFO	0x00800000	/* route information - RA	*/
+
 #define RTF_CACHE	0x01000000	/* cache entry			*/
 #define RTF_FLOW	0x02000000	/* flow significant route	*/
 #define RTF_POLICY	0x04000000	/* policy route			*/

include/net/ip6_route.h

@@ -7,6 +7,23 @@
 #define IP6_RT_PRIO_KERN	512
 #define IP6_RT_FLOW_MASK	0x00ff
 
+struct route_info {
+	__u8			type;
+	__u8			length;
+	__u8			prefix_len;
+#if defined(__BIG_ENDIAN_BITFIELD)
+	__u8			reserved_h:3,
+				route_pref:2,
+				reserved_l:3;
+#elif defined(__LITTLE_ENDIAN_BITFIELD)
+	__u8			reserved_l:3,
+				route_pref:2,
+				reserved_h:3;
+#endif
+	__u32			lifetime;
+	__u8			prefix[0];	/* 0,8 or 16 */
+};
+
 #ifdef __KERNEL__
 
 #include <net/flow.h>
@@ -92,6 +109,10 @@ extern struct rt6_info *	rt6_add_dflt_router(struct in6_addr *gwaddr,
 
 extern void			rt6_purge_dflt_routers(void);
 
+extern int			rt6_route_rcv(struct net_device *dev,
+					      u8 *opt, int len,
+					      struct in6_addr *gwaddr);
+
 extern void			rt6_redirect(struct in6_addr *dest,
 					     struct in6_addr *saddr,
 					     struct neighbour *neigh,

include/net/ndisc.h

@@ -22,6 +22,8 @@ enum {
 	ND_OPT_PREFIX_INFO = 3,		/* RFC2461 */
 	ND_OPT_REDIRECT_HDR = 4,	/* RFC2461 */
 	ND_OPT_MTU = 5,			/* RFC2461 */
+	__ND_OPT_ARRAY_MAX,
+	ND_OPT_ROUTE_INFO = 24,		/* RFC4191 */
 	__ND_OPT_MAX
 };
 

net/ipv6/Kconfig

@@ -49,6 +49,14 @@ config IPV6_ROUTER_PREF
 
 	  If unsure, say N.
 
+config IPV6_ROUTE_INFO
+	bool "IPv6: Route Information (RFC 4191) support (EXPERIMENTAL)"
+	depends on IPV6_ROUTER_PREF && EXPERIMENTAL
+	---help---
+	  This is experimental support of Route Information.
+
+	  If unsure, say N.
+
 config INET6_AH
 	tristate "IPv6: AH transformation"
 	depends on IPV6

net/ipv6/ndisc.c

@@ -156,7 +156,11 @@ struct neigh_table nd_tbl = {
 
 /* ND options */
 struct ndisc_options {
-	struct nd_opt_hdr *nd_opt_array[__ND_OPT_MAX];
+	struct nd_opt_hdr *nd_opt_array[__ND_OPT_ARRAY_MAX];
+#ifdef CONFIG_IPV6_ROUTE_INFO
+	struct nd_opt_hdr *nd_opts_ri;
+	struct nd_opt_hdr *nd_opts_ri_end;
+#endif
 };
 
 #define nd_opts_src_lladdr	nd_opt_array[ND_OPT_SOURCE_LL_ADDR]
@@ -255,6 +259,13 @@ static struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len,
 			if (ndopts->nd_opt_array[nd_opt->nd_opt_type] == 0)
 				ndopts->nd_opt_array[nd_opt->nd_opt_type] = nd_opt;
 			break;
+#ifdef CONFIG_IPV6_ROUTE_INFO
+		case ND_OPT_ROUTE_INFO:
+			ndopts->nd_opts_ri_end = nd_opt;
+			if (!ndopts->nd_opts_ri)
+				ndopts->nd_opts_ri = nd_opt;
+			break;
+#endif
 		default:
 			/*
 			 * Unknown options must be silently ignored,
@@ -1202,6 +1213,18 @@ skip_defrtr:
 			     NEIGH_UPDATE_F_ISROUTER);
 	}
 
+#ifdef CONFIG_IPV6_ROUTE_INFO
+	if (ndopts.nd_opts_ri) {
+		struct nd_opt_hdr *p;
+		for (p = ndopts.nd_opts_ri;
+		     p;
+		     p = ndisc_next_option(p, ndopts.nd_opts_ri_end)) {
+			rt6_route_rcv(skb->dev, (u8*)p, (p->nd_opt_len) << 3,
+				      &skb->nh.ipv6h->saddr);
+		}
+	}
+#endif
+
 	if (in6_dev->cnf.accept_ra_pinfo && ndopts.nd_opts_pi) {
 		struct nd_opt_hdr *p;
 		for (p = ndopts.nd_opts_pi;

net/ipv6/route.c

@@ -98,6 +98,14 @@ static int		ip6_pkt_discard_out(struct sk_buff *skb);
 static void		ip6_link_failure(struct sk_buff *skb);
 static void		ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu);
 
+#ifdef CONFIG_IPV6_ROUTE_INFO
+static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixlen,
+					   struct in6_addr *gwaddr, int ifindex,
+					   unsigned pref);
+static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixlen,
+					   struct in6_addr *gwaddr, int ifindex);
+#endif
+
 static struct dst_ops ip6_dst_ops = {
 	.family			=	AF_INET6,
 	.protocol		=	__constant_htons(ETH_P_IPV6),
@@ -346,6 +354,84 @@ static struct rt6_info *rt6_select(struct rt6_info **head, int oif,
 	return (match ? match : &ip6_null_entry);
 }
 
+#ifdef CONFIG_IPV6_ROUTE_INFO
+int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
+		  struct in6_addr *gwaddr)
+{
+	struct route_info *rinfo = (struct route_info *) opt;
+	struct in6_addr prefix_buf, *prefix;
+	unsigned int pref;
+	u32 lifetime;
+	struct rt6_info *rt;
+
+	if (len < sizeof(struct route_info)) {
+		return -EINVAL;
+	}
+
+	/* Sanity check for prefix_len and length */
+	if (rinfo->length > 3) {
+		return -EINVAL;
+	} else if (rinfo->prefix_len > 128) {
+		return -EINVAL;
+	} else if (rinfo->prefix_len > 64) {
+		if (rinfo->length < 2) {
+			return -EINVAL;
+		}
+	} else if (rinfo->prefix_len > 0) {
+		if (rinfo->length < 1) {
+			return -EINVAL;
+		}
+	}
+
+	pref = rinfo->route_pref;
+	if (pref == ICMPV6_ROUTER_PREF_INVALID)
+		pref = ICMPV6_ROUTER_PREF_MEDIUM;
+
+	lifetime = htonl(rinfo->lifetime);
+	if (lifetime == 0xffffffff) {
+		/* infinity */
+	} else if (lifetime > 0x7fffffff/HZ) {
+		/* Avoid arithmetic overflow */
+		lifetime = 0x7fffffff/HZ - 1;
+	}
+
+	if (rinfo->length == 3)
+		prefix = (struct in6_addr *)rinfo->prefix;
+	else {
+		/* this function is safe */
+		ipv6_addr_prefix(&prefix_buf,
+				 (struct in6_addr *)rinfo->prefix,
+				 rinfo->prefix_len);
+		prefix = &prefix_buf;
+	}
+
+	rt = rt6_get_route_info(prefix, rinfo->prefix_len, gwaddr, dev->ifindex);
+
+	if (rt && !lifetime) {
+		ip6_del_rt(rt, NULL, NULL, NULL);
+		rt = NULL;
+	}
+
+	if (!rt && lifetime)
+		rt = rt6_add_route_info(prefix, rinfo->prefix_len, gwaddr, dev->ifindex,
+					pref);
+	else if (rt)
+		rt->rt6i_flags = RTF_ROUTEINFO |
+				 (rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref);
+
+	if (rt) {
+		if (lifetime == 0xffffffff) {
+			rt->rt6i_flags &= ~RTF_EXPIRES;
+		} else {
+			rt->rt6i_expires = jiffies + HZ * lifetime;
+			rt->rt6i_flags |= RTF_EXPIRES;
+		}
+		dst_release(&rt->u.dst);
+	}
+	return 0;
+}
+#endif
+
 struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr,
 			    int oif, int strict)
 {
@@ -1277,6 +1363,54 @@ static struct rt6_info * ip6_rt_copy(struct rt6_info *ort)
 	return rt;
 }
 
+#ifdef CONFIG_IPV6_ROUTE_INFO
+static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixlen,
+					   struct in6_addr *gwaddr, int ifindex)
+{
+	struct fib6_node *fn;
+	struct rt6_info *rt = NULL;
+
+	write_lock_bh(&rt6_lock);
+	fn = fib6_locate(&ip6_routing_table, prefix ,prefixlen, NULL, 0);
+	if (!fn)
+		goto out;
+
+	for (rt = fn->leaf; rt; rt = rt->u.next) {
+		if (rt->rt6i_dev->ifindex != ifindex)
+			continue;
+		if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
+			continue;
+		if (!ipv6_addr_equal(&rt->rt6i_gateway, gwaddr))
+			continue;
+		dst_hold(&rt->u.dst);
+		break;
+	}
+out:
+	write_unlock_bh(&rt6_lock);
+	return rt;
+}
+
+static struct rt6_info *rt6_add_route_info(struct in6_addr *prefix, int prefixlen,
+					   struct in6_addr *gwaddr, int ifindex,
+					   unsigned pref)
+{
+	struct in6_rtmsg rtmsg;
+
+	memset(&rtmsg, 0, sizeof(rtmsg));
+	rtmsg.rtmsg_type = RTMSG_NEWROUTE;
+	ipv6_addr_copy(&rtmsg.rtmsg_dst, prefix);
+	rtmsg.rtmsg_dst_len = prefixlen;
+	ipv6_addr_copy(&rtmsg.rtmsg_gateway, gwaddr);
+	rtmsg.rtmsg_metric = 1024;
+	rtmsg.rtmsg_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO | RTF_UP | RTF_PREF(pref);
+	rtmsg.rtmsg_ifindex = ifindex;
+
+	ip6_route_add(&rtmsg, NULL, NULL, NULL);
+
+	return rt6_get_route_info(prefix, prefixlen, gwaddr, ifindex);
+}
+#endif
+
 struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *dev)
 {	
 	struct rt6_info *rt;