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

/*
 * net/dst.h	Protocol independent destination cache definitions.
 *
 * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
 *
 */

#ifndef _NET_DST_H
#define _NET_DST_H

#include <linux/netdevice.h>
#include <linux/rtnetlink.h>
#include <linux/rcupdate.h>
#include <linux/jiffies.h>
#include <net/neighbour.h>
#include <asm/processor.h>

/*
 * 0 - no debugging messages
 * 1 - rare events and bugs (default)
 * 2 - trace mode.
 */
#define RT_CACHE_DEBUG		0

#define DST_GC_MIN	(HZ/10)
#define DST_GC_INC	(HZ/2)
#define DST_GC_MAX	(120*HZ)

/* Each dst_entry has reference count and sits in some parent list(s).
 * When it is removed from parent list, it is "freed" (dst_free).
 * After this it enters dead state (dst->obsolete > 0) and if its refcnt
 * is zero, it can be destroyed immediately, otherwise it is added
 * to gc list and garbage collector periodically checks the refcnt.
 */

struct sk_buff;

struct dst_entry
{
	struct rcu_head		rcu_head;
	struct dst_entry	*child;
	struct net_device       *dev;
	short			error;
	short			obsolete;
	int			flags;
#define DST_HOST		1
#define DST_NOXFRM		2
#define DST_NOPOLICY		4
#define DST_NOHASH		8
	unsigned long		expires;

	unsigned short		header_len;	/* more space at head required */
	unsigned short		trailer_len;	/* space to reserve at tail */

	unsigned int		rate_tokens;
	unsigned long		rate_last;	/* rate limiting for ICMP */

	struct dst_entry	*path;

	struct neighbour	*neighbour;
	struct hh_cache		*hh;
#ifdef CONFIG_XFRM
	struct xfrm_state	*xfrm;
#else
	void			*__pad1;
#endif
	int			(*input)(struct sk_buff*);
	int			(*output)(struct sk_buff*);

	struct  dst_ops	        *ops;

	u32			metrics[RTAX_MAX];

#ifdef CONFIG_NET_CLS_ROUTE
	__u32			tclassid;
#else
	__u32			__pad2;
#endif


	/*
	 * Align __refcnt to a 64 bytes alignment
	 * (L1_CACHE_SIZE would be too much)
	 */
#ifdef CONFIG_64BIT
	long			__pad_to_align_refcnt[2];
#else
	long			__pad_to_align_refcnt[1];
#endif
	/*
	 * __refcnt wants to be on a different cache line from
	 * input/output/ops or performance tanks badly
	 */
	atomic_t		__refcnt;	/* client references	*/
	int			__use;
	unsigned long		lastuse;
	union {
		struct dst_entry *next;
		struct rtable    *rt_next;
		struct rt6_info   *rt6_next;
		struct dn_route  *dn_next;
	};
};


struct dst_ops
{
	unsigned short		family;
	__be16			protocol;
	unsigned		gc_thresh;

	int			(*gc)(struct dst_ops *ops);
	struct dst_entry *	(*check)(struct dst_entry *, __u32 cookie);
	void			(*destroy)(struct dst_entry *);
	void			(*ifdown)(struct dst_entry *,
					  struct net_device *dev, int how);
	struct dst_entry *	(*negative_advice)(struct dst_entry *);
	void			(*link_failure)(struct sk_buff *);
	void			(*update_pmtu)(struct dst_entry *dst, u32 mtu);
	int			(*local_out)(struct sk_buff *skb);

	atomic_t		entries;
	struct kmem_cache 		*kmem_cachep;
	struct net              *dst_net;
};

#ifdef __KERNEL__

static inline u32
dst_metric(const struct dst_entry *dst, int metric)
{
	return dst->metrics[metric-1];
}

static inline u32 dst_mtu(const struct dst_entry *dst)
{
	u32 mtu = dst_metric(dst, RTAX_MTU);
	/*
	 * Alexey put it here, so ask him about it :)
	 */
	barrier();
	return mtu;
}

/* RTT metrics are stored in milliseconds for user ABI, but used as jiffies */
static inline unsigned long dst_metric_rtt(const struct dst_entry *dst, int metric)
{
	return msecs_to_jiffies(dst_metric(dst, metric));
}

static inline void set_dst_metric_rtt(struct dst_entry *dst, int metric,
				      unsigned long rtt)
{
	dst->metrics[metric-1] = jiffies_to_msecs(rtt);
}

static inline u32
dst_allfrag(const struct dst_entry *dst)
{
	int ret = dst_metric(dst, RTAX_FEATURES) & RTAX_FEATURE_ALLFRAG;
	/* Yes, _exactly_. This is paranoia. */
	barrier();
	return ret;
}

static inline int
dst_metric_locked(struct dst_entry *dst, int metric)
{
	return dst_metric(dst, RTAX_LOCK) & (1<<metric);
}

static inline void dst_hold(struct dst_entry * dst)
{
	/*
	 * If your kernel compilation stops here, please check
	 * __pad_to_align_refcnt declaration in struct dst_entry
	 */
	BUILD_BUG_ON(offsetof(struct dst_entry, __refcnt) & 63);
	atomic_inc(&dst->__refcnt);
}

static inline void dst_use(struct dst_entry *dst, unsigned long time)
{
	dst_hold(dst);
	dst->__use++;
	dst->lastuse = time;
}

static inline
struct dst_entry * dst_clone(struct dst_entry * dst)
{
	if (dst)
		atomic_inc(&dst->__refcnt);
	return dst;
}

extern void dst_release(struct dst_entry *dst);
static inline void skb_dst_drop(struct sk_buff *skb)
{
	if (skb->_skb_dst)
		dst_release(skb_dst(skb));
	skb->_skb_dst = 0UL;
}

/* Children define the path of the packet through the
 * Linux networking.  Thus, destinations are stackable.
 */

static inline struct dst_entry *dst_pop(struct dst_entry *dst)
{
	struct dst_entry *child = dst_clone(dst->child);

	dst_release(dst);
	return child;
}

extern int dst_discard(struct sk_buff *skb);
extern void * dst_alloc(struct dst_ops * ops);
extern void __dst_free(struct dst_entry * dst);
extern struct dst_entry *dst_destroy(struct dst_entry * dst);

static inline void dst_free(struct dst_entry * dst)
{
	if (dst->obsolete > 1)
		return;
	if (!atomic_read(&dst->__refcnt)) {
		dst = dst_destroy(dst);
		if (!dst)
			return;
	}
	__dst_free(dst);
}

static inline void dst_rcu_free(struct rcu_head *head)
{
	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
	dst_free(dst);
}

static inline void dst_confirm(struct dst_entry *dst)
{
	if (dst)
		neigh_confirm(dst->neighbour);
}

static inline void dst_negative_advice(struct dst_entry **dst_p)
{
	struct dst_entry * dst = *dst_p;
	if (dst && dst->ops->negative_advice)
		*dst_p = dst->ops->negative_advice(dst);
}

static inline void dst_link_failure(struct sk_buff *skb)
{
	struct dst_entry *dst = skb_dst(skb);
	if (dst && dst->ops && dst->ops->link_failure)
		dst->ops->link_failure(skb);
}

static inline void dst_set_expires(struct dst_entry *dst, int timeout)
{
	unsigned long expires = jiffies + timeout;

	if (expires == 0)
		expires = 1;

	if (dst->expires == 0 || time_before(expires, dst->expires))
		dst->expires = expires;
}

/* Output packet to network from transport.  */
static inline int dst_output(struct sk_buff *skb)
{
	return skb_dst(skb)->output(skb);
}

/* Input packet from network to transport.  */
static inline int dst_input(struct sk_buff *skb)
{
	return skb_dst(skb)->input(skb);
}

static inline struct dst_entry *dst_check(struct dst_entry *dst, u32 cookie)
{
	if (dst->obsolete)
		dst = dst->ops->check(dst, cookie);
	return dst;
}

extern void		dst_init(void);

/* Flags for xfrm_lookup flags argument. */
enum {
	XFRM_LOOKUP_WAIT = 1 << 0,
	XFRM_LOOKUP_ICMP = 1 << 1,
};

struct flowi;
#ifndef CONFIG_XFRM
static inline int xfrm_lookup(struct net *net, struct dst_entry **dst_p,
			      struct flowi *fl, struct sock *sk, int flags)
{
	return 0;
} 
static inline int __xfrm_lookup(struct net *net, struct dst_entry **dst_p,
				struct flowi *fl, struct sock *sk, int flags)
{
	return 0;
}
#else
extern int xfrm_lookup(struct net *net, struct dst_entry **dst_p,
		       struct flowi *fl, struct sock *sk, int flags);
extern int __xfrm_lookup(struct net *net, struct dst_entry **dst_p,
			 struct flowi *fl, struct sock *sk, int flags);
#endif
#endif

#endif /* _NET_DST_H */
Commit	Line	Data
1da177e4 LT	1	/*
	2	* net/dst.h Protocol independent destination cache definitions.
	3	*
	4	* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
	5	*
	6	*/
	7
	8	#ifndef _NET_DST_H
	9	#define _NET_DST_H
	10
14c85021	11	#include <linux/netdevice.h>
1da177e4 LT	12	#include <linux/rtnetlink.h>
	13	#include <linux/rcupdate.h>
	14	#include <linux/jiffies.h>
	15	#include <net/neighbour.h>
	16	#include <asm/processor.h>
	17
	18	/*
	19	* 0 - no debugging messages
	20	* 1 - rare events and bugs (default)
	21	* 2 - trace mode.
	22	*/
	23	#define RT_CACHE_DEBUG 0
	24
	25	#define DST_GC_MIN (HZ/10)
	26	#define DST_GC_INC (HZ/2)
	27	#define DST_GC_MAX (120*HZ)
	28
	29	/* Each dst_entry has reference count and sits in some parent list(s).
	30	* When it is removed from parent list, it is "freed" (dst_free).
	31	* After this it enters dead state (dst->obsolete > 0) and if its refcnt
	32	* is zero, it can be destroyed immediately, otherwise it is added
	33	* to gc list and garbage collector periodically checks the refcnt.
	34	*/
	35
	36	struct sk_buff;
	37
	38	struct dst_entry
	39	{
1e19e02c	40	struct rcu_head rcu_head;
1da177e4 LT	41	struct dst_entry *child;
1da177e4 LT	42	struct net_device *dev;
c4d54110 HX	43	short error;
c4d54110 HX	44	short obsolete;
1da177e4 LT	45	int flags;
	46	#define DST_HOST 1
	47	#define DST_NOXFRM 2
	48	#define DST_NOPOLICY 4
	49	#define DST_NOHASH 8
1da177e4 LT	50	unsigned long expires;
	51
	52	unsigned short header_len; /* more space at head required */
	53	unsigned short trailer_len; /* space to reserve at tail */
	54
69a73829	55	unsigned int rate_tokens;
f1dd9c37	56	unsigned long rate_last; /* rate limiting for ICMP */
69a73829	57
f1dd9c37	58	struct dst_entry *path;
1da177e4	59
1da177e4 LT	60	struct neighbour *neighbour;
1da177e4 LT	61	struct hh_cache *hh;
def8b4fa	62	#ifdef CONFIG_XFRM
1da177e4	63	struct xfrm_state *xfrm;
5635c10d ED	64	#else
5635c10d ED	65	void *__pad1;
def8b4fa	66	#endif
1da177e4 LT	67	int (input)(struct sk_buff);
	68	int (output)(struct sk_buff);
	69
1da177e4	70	struct dst_ops *ops;
f1dd9c37 ZY	71
	72	u32 metrics[RTAX_MAX];
	73
	74	#ifdef CONFIG_NET_CLS_ROUTE
	75	__u32 tclassid;
5635c10d ED	76	#else
5635c10d ED	77	__u32 __pad2;
f1dd9c37 ZY	78	#endif
f1dd9c37 ZY	79
5635c10d ED	80
	81	/*
	82	* Align __refcnt to a 64 bytes alignment
	83	* (L1_CACHE_SIZE would be too much)
	84	*/
	85	#ifdef CONFIG_64BIT
	86	long __pad_to_align_refcnt[2];
	87	#else
	88	long __pad_to_align_refcnt[1];
	89	#endif
f1dd9c37 ZY	90	/*
	91	* __refcnt wants to be on a different cache line from
	92	* input/output/ops or performance tanks badly
	93	*/
1e19e02c ED	94	atomic_t __refcnt; /* client references */
1e19e02c ED	95	int __use;
f1dd9c37	96	unsigned long lastuse;
1e19e02c ED	97	union {
	98	struct dst_entry *next;
	99	struct rtable *rt_next;
	100	struct rt6_info *rt6_next;
	101	struct dn_route *dn_next;
	102	};
1da177e4 LT	103	};
	104
	105
	106	struct dst_ops
	107	{
	108	unsigned short family;
d77072ec	109	__be16 protocol;
1da177e4 LT	110	unsigned gc_thresh;
1da177e4 LT	111
569d3645	112	int (gc)(struct dst_ops ops);
1da177e4 LT	113	struct dst_entry * (check)(struct dst_entry , __u32 cookie);
	114	void (destroy)(struct dst_entry );
	115	void (ifdown)(struct dst_entry ,
	116	struct net_device *dev, int how);
	117	struct dst_entry * (negative_advice)(struct dst_entry );
	118	void (link_failure)(struct sk_buff );
	119	void (update_pmtu)(struct dst_entry dst, u32 mtu);
862b82c6	120	int (local_out)(struct sk_buff skb);
1da177e4 LT	121
1da177e4 LT	122	atomic_t entries;
e18b890b	123	struct kmem_cache *kmem_cachep;
d4fa26ff	124	struct net *dst_net;
1da177e4 LT	125	};
	126
	127	#ifdef __KERNEL__
	128
	129	static inline u32
	130	dst_metric(const struct dst_entry *dst, int metric)
	131	{
	132	return dst->metrics[metric-1];
	133	}
	134
	135	static inline u32 dst_mtu(const struct dst_entry *dst)
	136	{
	137	u32 mtu = dst_metric(dst, RTAX_MTU);
	138	/*
	139	* Alexey put it here, so ask him about it :)
	140	*/
	141	barrier();
	142	return mtu;
	143	}
	144
c1e20f7c SH	145	/* RTT metrics are stored in milliseconds for user ABI, but used as jiffies */
	146	static inline unsigned long dst_metric_rtt(const struct dst_entry *dst, int metric)
	147	{
	148	return msecs_to_jiffies(dst_metric(dst, metric));
	149	}
	150
	151	static inline void set_dst_metric_rtt(struct dst_entry *dst, int metric,
	152	unsigned long rtt)
	153	{
	154	dst->metrics[metric-1] = jiffies_to_msecs(rtt);
	155	}
	156
1da177e4 LT	157	static inline u32
	158	dst_allfrag(const struct dst_entry *dst)
	159	{
	160	int ret = dst_metric(dst, RTAX_FEATURES) & RTAX_FEATURE_ALLFRAG;
	161	/* Yes, _exactly_. This is paranoia. */
	162	barrier();
	163	return ret;
	164	}
	165
	166	static inline int
	167	dst_metric_locked(struct dst_entry *dst, int metric)
	168	{
	169	return dst_metric(dst, RTAX_LOCK) & (1<<metric);
	170	}
	171
	172	static inline void dst_hold(struct dst_entry * dst)
	173	{
5635c10d ED	174	/*
	175	* If your kernel compilation stops here, please check
	176	* __pad_to_align_refcnt declaration in struct dst_entry
	177	*/
	178	BUILD_BUG_ON(offsetof(struct dst_entry, __refcnt) & 63);
1da177e4 LT	179	atomic_inc(&dst->__refcnt);
	180	}
	181
03f49f34 PE	182	static inline void dst_use(struct dst_entry *dst, unsigned long time)
	183	{
	184	dst_hold(dst);
	185	dst->__use++;
	186	dst->lastuse = time;
	187	}
	188
1da177e4 LT	189	static inline
	190	struct dst_entry * dst_clone(struct dst_entry * dst)
	191	{
	192	if (dst)
	193	atomic_inc(&dst->__refcnt);
	194	return dst;
	195	}
	196
8d330868	197	extern void dst_release(struct dst_entry *dst);
adf30907 ED	198	static inline void skb_dst_drop(struct sk_buff *skb)
	199	{
	200	if (skb->_skb_dst)
	201	dst_release(skb_dst(skb));
	202	skb->_skb_dst = 0UL;
	203	}
1da177e4 LT	204
	205	/* Children define the path of the packet through the
	206	* Linux networking. Thus, destinations are stackable.
	207	*/
	208
	209	static inline struct dst_entry dst_pop(struct dst_entry dst)
	210	{
	211	struct dst_entry *child = dst_clone(dst->child);
	212
	213	dst_release(dst);
	214	return child;
	215	}
	216
352e512c	217	extern int dst_discard(struct sk_buff *skb);
1da177e4 LT	218	extern void * dst_alloc(struct dst_ops * ops);
	219	extern void __dst_free(struct dst_entry * dst);
	220	extern struct dst_entry dst_destroy(struct dst_entry dst);
	221
	222	static inline void dst_free(struct dst_entry * dst)
	223	{
	224	if (dst->obsolete > 1)
	225	return;
	226	if (!atomic_read(&dst->__refcnt)) {
	227	dst = dst_destroy(dst);
	228	if (!dst)
	229	return;
	230	}
	231	__dst_free(dst);
	232	}
	233
	234	static inline void dst_rcu_free(struct rcu_head *head)
	235	{
	236	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
	237	dst_free(dst);
	238	}
	239
	240	static inline void dst_confirm(struct dst_entry *dst)
	241	{
	242	if (dst)
	243	neigh_confirm(dst->neighbour);
	244	}
	245
	246	static inline void dst_negative_advice(struct dst_entry **dst_p)
	247	{
	248	struct dst_entry * dst = *dst_p;
	249	if (dst && dst->ops->negative_advice)
	250	*dst_p = dst->ops->negative_advice(dst);
	251	}
	252
	253	static inline void dst_link_failure(struct sk_buff *skb)
	254	{
adf30907	255	struct dst_entry *dst = skb_dst(skb);
1da177e4 LT	256	if (dst && dst->ops && dst->ops->link_failure)
	257	dst->ops->link_failure(skb);
	258	}
	259
	260	static inline void dst_set_expires(struct dst_entry *dst, int timeout)
	261	{
	262	unsigned long expires = jiffies + timeout;
	263
	264	if (expires == 0)
	265	expires = 1;
	266
	267	if (dst->expires == 0 \|\| time_before(expires, dst->expires))
	268	dst->expires = expires;
	269	}
	270
	271	/* Output packet to network from transport. */
	272	static inline int dst_output(struct sk_buff *skb)
	273	{
adf30907	274	return skb_dst(skb)->output(skb);
1da177e4 LT	275	}
	276
	277	/* Input packet from network to transport. */
	278	static inline int dst_input(struct sk_buff *skb)
	279	{
adf30907	280	return skb_dst(skb)->input(skb);
1da177e4 LT	281	}
	282
	283	static inline struct dst_entry dst_check(struct dst_entry dst, u32 cookie)
	284	{
	285	if (dst->obsolete)
	286	dst = dst->ops->check(dst, cookie);
	287	return dst;
	288	}
	289
	290	extern void dst_init(void);
	291
815f4e57 HX	292	/* Flags for xfrm_lookup flags argument. */
	293	enum {
	294	XFRM_LOOKUP_WAIT = 1 << 0,
8b7817f3	295	XFRM_LOOKUP_ICMP = 1 << 1,
815f4e57 HX	296	};
815f4e57 HX	297
1da177e4 LT	298	struct flowi;
1da177e4 LT	299	#ifndef CONFIG_XFRM
52479b62 AD	300	static inline int xfrm_lookup(struct net net, struct dst_entry *dst_p,
52479b62 AD	301	struct flowi fl, struct sock sk, int flags)
1da177e4 LT	302	{
	303	return 0;
	304	}
52479b62 AD	305	static inline int __xfrm_lookup(struct net net, struct dst_entry *dst_p,
52479b62 AD	306	struct flowi fl, struct sock sk, int flags)
14e50e57 DM	307	{
	308	return 0;
	309	}
1da177e4	310	#else
52479b62 AD	311	extern int xfrm_lookup(struct net net, struct dst_entry *dst_p,
	312	struct flowi fl, struct sock sk, int flags);
	313	extern int __xfrm_lookup(struct net net, struct dst_entry *dst_p,
	314	struct flowi fl, struct sock sk, int flags);
1da177e4 LT	315	#endif
	316	#endif
	317
	318	#endif /* _NET_DST_H */