[deliverable/linux.git] / include / net / route.h

/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET  is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		Definitions for the IP router.
 *
 * Version:	@(#)route.h	1.0.4	05/27/93
 *
 * Authors:	Ross Biro
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 * Fixes:
 *		Alan Cox	:	Reformatted. Added ip_rt_local()
 *		Alan Cox	:	Support for TCP parameters.
 *		Alexey Kuznetsov:	Major changes for new routing code.
 *		Mike McLagan    :	Routing by source
 *		Robert Olsson   :	Added rt_cache statistics
 *
 *		This program is free software; you can redistribute it and/or
 *		modify it under the terms of the GNU General Public License
 *		as published by the Free Software Foundation; either version
 *		2 of the License, or (at your option) any later version.
 */
#ifndef _ROUTE_H
#define _ROUTE_H

#include <net/dst.h>
#include <net/inetpeer.h>
#include <net/flow.h>
#include <net/inet_sock.h>
#include <linux/in_route.h>
#include <linux/rtnetlink.h>
#include <linux/route.h>
#include <linux/ip.h>
#include <linux/cache.h>
#include <linux/security.h>

#define RTO_ONLINK	0x01

#define RT_CONN_FLAGS(sk)   (RT_TOS(inet_sk(sk)->tos) | sock_flag(sk, SOCK_LOCALROUTE))

struct fib_nh;
struct inet_peer;
struct fib_info;
struct rtable {
	struct dst_entry	dst;

	/* Lookup key. */
	__be32			rt_key_dst;
	__be32			rt_key_src;

	int			rt_genid;
	unsigned		rt_flags;
	__u16			rt_type;
	__u8			rt_key_tos;

	__be32			rt_dst;	/* Path destination	*/
	__be32			rt_src;	/* Path source		*/
	int			rt_route_iif;
	int			rt_iif;
	int			rt_oif;
	__u32			rt_mark;

	/* Info on neighbour */
	__be32			rt_gateway;

	/* Miscellaneous cached information */
	__be32			rt_spec_dst; /* RFC1122 specific destination */
	u32			rt_peer_genid;
	struct inet_peer	*peer; /* long-living peer info */
	struct fib_info		*fi; /* for client ref to shared metrics */
};

static inline bool rt_is_input_route(struct rtable *rt)
{
	return rt->rt_route_iif != 0;
}

static inline bool rt_is_output_route(struct rtable *rt)
{
	return rt->rt_route_iif == 0;
}

struct ip_rt_acct {
	__u32 	o_bytes;
	__u32 	o_packets;
	__u32 	i_bytes;
	__u32 	i_packets;
};

struct rt_cache_stat {
        unsigned int in_hit;
        unsigned int in_slow_tot;
        unsigned int in_slow_mc;
        unsigned int in_no_route;
        unsigned int in_brd;
        unsigned int in_martian_dst;
        unsigned int in_martian_src;
        unsigned int out_hit;
        unsigned int out_slow_tot;
        unsigned int out_slow_mc;
        unsigned int gc_total;
        unsigned int gc_ignored;
        unsigned int gc_goal_miss;
        unsigned int gc_dst_overflow;
        unsigned int in_hlist_search;
        unsigned int out_hlist_search;
};

extern struct ip_rt_acct __percpu *ip_rt_acct;

struct in_device;
extern int		ip_rt_init(void);
extern void		ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
				       __be32 src, struct net_device *dev);
extern void		rt_cache_flush(struct net *net, int how);
extern void		rt_cache_flush_batch(struct net *net);
extern struct rtable *__ip_route_output_key(struct net *, struct flowi4 *flp);
extern struct rtable *ip_route_output_flow(struct net *, struct flowi4 *flp,
					   struct sock *sk);
extern struct dst_entry *ipv4_blackhole_route(struct net *net, struct dst_entry *dst_orig);

static inline struct rtable *ip_route_output_key(struct net *net, struct flowi4 *flp)
{
	return ip_route_output_flow(net, flp, NULL);
}

static inline struct rtable *ip_route_output(struct net *net, __be32 daddr,
					     __be32 saddr, u8 tos, int oif)
{
	struct flowi4 fl4 = {
		.flowi4_oif = oif,
		.daddr = daddr,
		.saddr = saddr,
		.flowi4_tos = tos,
	};
	return ip_route_output_key(net, &fl4);
}

static inline struct rtable *ip_route_output_ports(struct net *net, struct sock *sk,
						   __be32 daddr, __be32 saddr,
						   __be16 dport, __be16 sport,
						   __u8 proto, __u8 tos, int oif)
{
	struct flowi4 fl4;

	flowi4_init_output(&fl4, oif, sk ? sk->sk_mark : 0, tos,
			   RT_SCOPE_UNIVERSE, proto,
			   sk ? inet_sk_flowi_flags(sk) : 0,
			   daddr, saddr, dport, sport);
	if (sk)
		security_sk_classify_flow(sk, flowi4_to_flowi(&fl4));
	return ip_route_output_flow(net, &fl4, sk);
}

static inline struct rtable *ip_route_output_gre(struct net *net,
						 __be32 daddr, __be32 saddr,
						 __be32 gre_key, __u8 tos, int oif)
{
	struct flowi4 fl4 = {
		.flowi4_oif = oif,
		.daddr = daddr,
		.saddr = saddr,
		.flowi4_tos = tos,
		.flowi4_proto = IPPROTO_GRE,
		.fl4_gre_key = gre_key,
	};
	return ip_route_output_key(net, &fl4);
}

extern int ip_route_input_common(struct sk_buff *skb, __be32 dst, __be32 src,
				 u8 tos, struct net_device *devin, bool noref);

static inline int ip_route_input(struct sk_buff *skb, __be32 dst, __be32 src,
				 u8 tos, struct net_device *devin)
{
	return ip_route_input_common(skb, dst, src, tos, devin, false);
}

static inline int ip_route_input_noref(struct sk_buff *skb, __be32 dst, __be32 src,
				       u8 tos, struct net_device *devin)
{
	return ip_route_input_common(skb, dst, src, tos, devin, true);
}

extern unsigned short	ip_rt_frag_needed(struct net *net, const struct iphdr *iph,
					  unsigned short new_mtu, struct net_device *dev);
extern void		ip_rt_send_redirect(struct sk_buff *skb);

extern unsigned		inet_addr_type(struct net *net, __be32 addr);
extern unsigned		inet_dev_addr_type(struct net *net, const struct net_device *dev, __be32 addr);
extern void		ip_rt_multicast_event(struct in_device *);
extern int		ip_rt_ioctl(struct net *, unsigned int cmd, void __user *arg);
extern void		ip_rt_get_source(u8 *src, struct rtable *rt);
extern int		ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb);

struct in_ifaddr;
extern void fib_add_ifaddr(struct in_ifaddr *);
extern void fib_del_ifaddr(struct in_ifaddr *, struct in_ifaddr *);

static inline void ip_rt_put(struct rtable * rt)
{
	if (rt)
		dst_release(&rt->dst);
}

#define IPTOS_RT_MASK	(IPTOS_TOS_MASK & ~3)

extern const __u8 ip_tos2prio[16];

static inline char rt_tos2priority(u8 tos)
{
	return ip_tos2prio[IPTOS_TOS(tos)>>1];
}

/* ip_route_connect() and ip_route_newports() work in tandem whilst
 * binding a socket for a new outgoing connection.
 *
 * In order to use IPSEC properly, we must, in the end, have a
 * route that was looked up using all available keys including source
 * and destination ports.
 *
 * However, if a source port needs to be allocated (the user specified
 * a wildcard source port) we need to obtain addressing information
 * in order to perform that allocation.
 *
 * So ip_route_connect() looks up a route using wildcarded source and
 * destination ports in the key, simply so that we can get a pair of
 * addresses to use for port allocation.
 *
 * Later, once the ports are allocated, ip_route_newports() will make
 * another route lookup if needed to make sure we catch any IPSEC
 * rules keyed on the port information.
 *
 * The callers allocate the flow key on their stack, and must pass in
 * the same flowi4 object to both the ip_route_connect() and the
 * ip_route_newports() calls.
 */

static inline void ip_route_connect_init(struct flowi4 *fl4, __be32 dst, __be32 src,
					 u32 tos, int oif, u8 protocol,
					 __be16 sport, __be16 dport,
					 struct sock *sk, bool can_sleep)
{
	__u8 flow_flags = 0;

	if (inet_sk(sk)->transparent)
		flow_flags |= FLOWI_FLAG_ANYSRC;
	if (protocol == IPPROTO_TCP)
		flow_flags |= FLOWI_FLAG_PRECOW_METRICS;
	if (can_sleep)
		flow_flags |= FLOWI_FLAG_CAN_SLEEP;

	flowi4_init_output(fl4, oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE,
			   protocol, flow_flags, dst, src, dport, sport);
}

static inline struct rtable *ip_route_connect(struct flowi4 *fl4,
					      __be32 dst, __be32 src, u32 tos,
					      int oif, u8 protocol,
					      __be16 sport, __be16 dport,
					      struct sock *sk, bool can_sleep)
{
	struct net *net = sock_net(sk);
	struct rtable *rt;

	ip_route_connect_init(fl4, dst, src, tos, oif, protocol,
			      sport, dport, sk, can_sleep);

	if (!dst || !src) {
		rt = __ip_route_output_key(net, fl4);
		if (IS_ERR(rt))
			return rt;
		ip_rt_put(rt);
	}
	security_sk_classify_flow(sk, flowi4_to_flowi(fl4));
	return ip_route_output_flow(net, fl4, sk);
}

static inline struct rtable *ip_route_newports(struct flowi4 *fl4, struct rtable *rt,
					       __be16 orig_sport, __be16 orig_dport,
					       __be16 sport, __be16 dport,
					       struct sock *sk)
{
	if (sport != orig_sport || dport != orig_dport) {
		fl4->fl4_dport = dport;
		fl4->fl4_sport = sport;
		ip_rt_put(rt);
		security_sk_classify_flow(sk, flowi4_to_flowi(fl4));
		return ip_route_output_flow(sock_net(sk), fl4, sk);
	}
	return rt;
}

extern void rt_bind_peer(struct rtable *rt, int create);

static inline struct inet_peer *rt_get_peer(struct rtable *rt)
{
	if (rt->peer)
		return rt->peer;

	rt_bind_peer(rt, 0);
	return rt->peer;
}

static inline int inet_iif(const struct sk_buff *skb)
{
	return skb_rtable(skb)->rt_iif;
}

extern int sysctl_ip_default_ttl;

static inline int ip4_dst_hoplimit(const struct dst_entry *dst)
{
	int hoplimit = dst_metric_raw(dst, RTAX_HOPLIMIT);

	if (hoplimit == 0)
		hoplimit = sysctl_ip_default_ttl;
	return hoplimit;
}

#endif	/* _ROUTE_H */
Commit	Line	Data
	1	/*
	2	* INET An implementation of the TCP/IP protocol suite for the LINUX
	3	* operating system. INET is implemented using the BSD Socket
	4	* interface as the means of communication with the user level.
	5	*
	6	* Definitions for the IP router.
	7	*
	8	* Version: @(#)route.h 1.0.4 05/27/93
	9	*
	10	* Authors: Ross Biro
	11	* Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
	12	* Fixes:
	13	* Alan Cox : Reformatted. Added ip_rt_local()
	14	* Alan Cox : Support for TCP parameters.
	15	* Alexey Kuznetsov: Major changes for new routing code.
	16	* Mike McLagan : Routing by source
	17	* Robert Olsson : Added rt_cache statistics
	18	*
	19	* This program is free software; you can redistribute it and/or
	20	* modify it under the terms of the GNU General Public License
	21	* as published by the Free Software Foundation; either version
	22	* 2 of the License, or (at your option) any later version.
	23	*/
	24	#ifndef _ROUTE_H
	25	#define _ROUTE_H
	26
	27	#include <net/dst.h>
	28	#include <net/inetpeer.h>
	29	#include <net/flow.h>
	30	#include <net/inet_sock.h>
	31	#include <linux/in_route.h>
	32	#include <linux/rtnetlink.h>
	33	#include <linux/route.h>
	34	#include <linux/ip.h>
	35	#include <linux/cache.h>
	36	#include <linux/security.h>
	37
	38	#define RTO_ONLINK 0x01
	39
	40	#define RT_CONN_FLAGS(sk) (RT_TOS(inet_sk(sk)->tos) \| sock_flag(sk, SOCK_LOCALROUTE))
	41
	42	struct fib_nh;
	43	struct inet_peer;
	44	struct fib_info;
	45	struct rtable {
	46	struct dst_entry dst;
	47
	48	/* Lookup key. */
	49	__be32 rt_key_dst;
	50	__be32 rt_key_src;
	51
	52	int rt_genid;
	53	unsigned rt_flags;
	54	__u16 rt_type;
	55	__u8 rt_key_tos;
	56
	57	__be32 rt_dst; /* Path destination */
	58	__be32 rt_src; /* Path source */
	59	int rt_route_iif;
	60	int rt_iif;
	61	int rt_oif;
	62	__u32 rt_mark;
	63
	64	/* Info on neighbour */
	65	__be32 rt_gateway;
	66
	67	/* Miscellaneous cached information */
	68	__be32 rt_spec_dst; /* RFC1122 specific destination */
	69	u32 rt_peer_genid;
	70	struct inet_peer peer; / long-living peer info */
	71	struct fib_info fi; / for client ref to shared metrics */
	72	};
	73
	74	static inline bool rt_is_input_route(struct rtable *rt)
	75	{
	76	return rt->rt_route_iif != 0;
	77	}
	78
	79	static inline bool rt_is_output_route(struct rtable *rt)
	80	{
	81	return rt->rt_route_iif == 0;
	82	}
	83
	84	struct ip_rt_acct {
	85	__u32 o_bytes;
	86	__u32 o_packets;
	87	__u32 i_bytes;
	88	__u32 i_packets;
	89	};
	90
	91	struct rt_cache_stat {
	92	unsigned int in_hit;
	93	unsigned int in_slow_tot;
	94	unsigned int in_slow_mc;
	95	unsigned int in_no_route;
	96	unsigned int in_brd;
	97	unsigned int in_martian_dst;
	98	unsigned int in_martian_src;
	99	unsigned int out_hit;
	100	unsigned int out_slow_tot;
	101	unsigned int out_slow_mc;
	102	unsigned int gc_total;
	103	unsigned int gc_ignored;
	104	unsigned int gc_goal_miss;
	105	unsigned int gc_dst_overflow;
	106	unsigned int in_hlist_search;
	107	unsigned int out_hlist_search;
	108	};
	109
	110	extern struct ip_rt_acct __percpu *ip_rt_acct;
	111
	112	struct in_device;
	113	extern int ip_rt_init(void);
	114	extern void ip_rt_redirect(__be32 old_gw, __be32 dst, __be32 new_gw,
	115	__be32 src, struct net_device *dev);
	116	extern void rt_cache_flush(struct net *net, int how);
	117	extern void rt_cache_flush_batch(struct net *net);
	118	extern struct rtable __ip_route_output_key(struct net , struct flowi4 *flp);
	119	extern struct rtable ip_route_output_flow(struct net , struct flowi4 *flp,
	120	struct sock *sk);
	121	extern struct dst_entry ipv4_blackhole_route(struct net net, struct dst_entry *dst_orig);
	122
	123	static inline struct rtable ip_route_output_key(struct net net, struct flowi4 *flp)
	124	{
	125	return ip_route_output_flow(net, flp, NULL);
	126	}
	127
	128	static inline struct rtable ip_route_output(struct net net, __be32 daddr,
	129	__be32 saddr, u8 tos, int oif)
	130	{
	131	struct flowi4 fl4 = {
	132	.flowi4_oif = oif,
	133	.daddr = daddr,
	134	.saddr = saddr,
	135	.flowi4_tos = tos,
	136	};
	137	return ip_route_output_key(net, &fl4);
	138	}
	139
	140	static inline struct rtable ip_route_output_ports(struct net net, struct sock *sk,
	141	__be32 daddr, __be32 saddr,
	142	__be16 dport, __be16 sport,
	143	__u8 proto, __u8 tos, int oif)
	144	{
	145	struct flowi4 fl4;
	146
	147	flowi4_init_output(&fl4, oif, sk ? sk->sk_mark : 0, tos,
	148	RT_SCOPE_UNIVERSE, proto,
	149	sk ? inet_sk_flowi_flags(sk) : 0,
	150	daddr, saddr, dport, sport);
	151	if (sk)
	152	security_sk_classify_flow(sk, flowi4_to_flowi(&fl4));
	153	return ip_route_output_flow(net, &fl4, sk);
	154	}
	155
	156	static inline struct rtable ip_route_output_gre(struct net net,
	157	__be32 daddr, __be32 saddr,
	158	__be32 gre_key, __u8 tos, int oif)
	159	{
	160	struct flowi4 fl4 = {
	161	.flowi4_oif = oif,
	162	.daddr = daddr,
	163	.saddr = saddr,
	164	.flowi4_tos = tos,
	165	.flowi4_proto = IPPROTO_GRE,
	166	.fl4_gre_key = gre_key,
	167	};
	168	return ip_route_output_key(net, &fl4);
	169	}
	170
	171	extern int ip_route_input_common(struct sk_buff *skb, __be32 dst, __be32 src,
	172	u8 tos, struct net_device *devin, bool noref);
	173
	174	static inline int ip_route_input(struct sk_buff *skb, __be32 dst, __be32 src,
	175	u8 tos, struct net_device *devin)
	176	{
	177	return ip_route_input_common(skb, dst, src, tos, devin, false);
	178	}
	179
	180	static inline int ip_route_input_noref(struct sk_buff *skb, __be32 dst, __be32 src,
	181	u8 tos, struct net_device *devin)
	182	{
	183	return ip_route_input_common(skb, dst, src, tos, devin, true);
	184	}
	185
	186	extern unsigned short ip_rt_frag_needed(struct net net, const struct iphdr iph,
	187	unsigned short new_mtu, struct net_device *dev);
	188	extern void ip_rt_send_redirect(struct sk_buff *skb);
	189
	190	extern unsigned inet_addr_type(struct net *net, __be32 addr);
	191	extern unsigned inet_dev_addr_type(struct net net, const struct net_device dev, __be32 addr);
	192	extern void ip_rt_multicast_event(struct in_device *);
	193	extern int ip_rt_ioctl(struct net , unsigned int cmd, void __user arg);
	194	extern void ip_rt_get_source(u8 src, struct rtable rt);
	195	extern int ip_rt_dump(struct sk_buff skb, struct netlink_callback cb);
	196
	197	struct in_ifaddr;
	198	extern void fib_add_ifaddr(struct in_ifaddr *);
	199	extern void fib_del_ifaddr(struct in_ifaddr , struct in_ifaddr );
	200
	201	static inline void ip_rt_put(struct rtable * rt)
	202	{
	203	if (rt)
	204	dst_release(&rt->dst);
	205	}
	206
	207	#define IPTOS_RT_MASK (IPTOS_TOS_MASK & ~3)
	208
	209	extern const __u8 ip_tos2prio[16];
	210
	211	static inline char rt_tos2priority(u8 tos)
	212	{
	213	return ip_tos2prio[IPTOS_TOS(tos)>>1];
	214	}
	215
	216	/* ip_route_connect() and ip_route_newports() work in tandem whilst
	217	* binding a socket for a new outgoing connection.
	218	*
	219	* In order to use IPSEC properly, we must, in the end, have a
	220	* route that was looked up using all available keys including source
	221	* and destination ports.
	222	*
	223	* However, if a source port needs to be allocated (the user specified
	224	* a wildcard source port) we need to obtain addressing information
	225	* in order to perform that allocation.
	226	*
	227	* So ip_route_connect() looks up a route using wildcarded source and
	228	* destination ports in the key, simply so that we can get a pair of
	229	* addresses to use for port allocation.
	230	*
	231	* Later, once the ports are allocated, ip_route_newports() will make
	232	* another route lookup if needed to make sure we catch any IPSEC
	233	* rules keyed on the port information.
	234	*
	235	* The callers allocate the flow key on their stack, and must pass in
	236	* the same flowi4 object to both the ip_route_connect() and the
	237	* ip_route_newports() calls.
	238	*/
	239
	240	static inline void ip_route_connect_init(struct flowi4 *fl4, __be32 dst, __be32 src,
	241	u32 tos, int oif, u8 protocol,
	242	__be16 sport, __be16 dport,
	243	struct sock *sk, bool can_sleep)
	244	{
	245	__u8 flow_flags = 0;
	246
	247	if (inet_sk(sk)->transparent)
	248	flow_flags \|= FLOWI_FLAG_ANYSRC;
	249	if (protocol == IPPROTO_TCP)
	250	flow_flags \|= FLOWI_FLAG_PRECOW_METRICS;
	251	if (can_sleep)
	252	flow_flags \|= FLOWI_FLAG_CAN_SLEEP;
	253
	254	flowi4_init_output(fl4, oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE,
	255	protocol, flow_flags, dst, src, dport, sport);
	256	}
	257
	258	static inline struct rtable ip_route_connect(struct flowi4 fl4,
	259	__be32 dst, __be32 src, u32 tos,
	260	int oif, u8 protocol,
	261	__be16 sport, __be16 dport,
	262	struct sock *sk, bool can_sleep)
	263	{
	264	struct net *net = sock_net(sk);
	265	struct rtable *rt;
	266
	267	ip_route_connect_init(fl4, dst, src, tos, oif, protocol,
	268	sport, dport, sk, can_sleep);
	269
	270	if (!dst \|\| !src) {
	271	rt = __ip_route_output_key(net, fl4);
	272	if (IS_ERR(rt))
	273	return rt;
	274	ip_rt_put(rt);
	275	}
	276	security_sk_classify_flow(sk, flowi4_to_flowi(fl4));
	277	return ip_route_output_flow(net, fl4, sk);
	278	}
	279
	280	static inline struct rtable ip_route_newports(struct flowi4 fl4, struct rtable *rt,
	281	__be16 orig_sport, __be16 orig_dport,
	282	__be16 sport, __be16 dport,
	283	struct sock *sk)
	284	{
	285	if (sport != orig_sport \|\| dport != orig_dport) {
	286	fl4->fl4_dport = dport;
	287	fl4->fl4_sport = sport;
	288	ip_rt_put(rt);
	289	security_sk_classify_flow(sk, flowi4_to_flowi(fl4));
	290	return ip_route_output_flow(sock_net(sk), fl4, sk);
	291	}
	292	return rt;
	293	}
	294
	295	extern void rt_bind_peer(struct rtable *rt, int create);
	296
	297	static inline struct inet_peer rt_get_peer(struct rtable rt)
	298	{
	299	if (rt->peer)
	300	return rt->peer;
	301
	302	rt_bind_peer(rt, 0);
	303	return rt->peer;
	304	}
	305
	306	static inline int inet_iif(const struct sk_buff *skb)
	307	{
	308	return skb_rtable(skb)->rt_iif;
	309	}
	310
	311	extern int sysctl_ip_default_ttl;
	312
	313	static inline int ip4_dst_hoplimit(const struct dst_entry *dst)
	314	{
	315	int hoplimit = dst_metric_raw(dst, RTAX_HOPLIMIT);
	316
	317	if (hoplimit == 0)
	318	hoplimit = sysctl_ip_default_ttl;
	319	return hoplimit;
	320	}
	321
	322	#endif /* _ROUTE_H */