[net-next-2.6.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 <net/dst_ops.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;
	};
};

#ifdef __KERNEL__

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

static inline u32
dst_feature(const struct dst_entry *dst, u32 feature)
{
	return dst_metric(dst, RTAX_FEATURES) & feature;
}

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_feature(dst,  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 sock *sk)
{
	struct dst_entry * dst = *dst_p;
	if (dst && dst->ops->negative_advice) {
		*dst_p = dst->ops->negative_advice(dst);

		if (dst != *dst_p) {
			extern void sk_reset_txq(struct sock *sk);

			sk_reset_txq(sk);
		}
	}
}

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
86393e52	11	#include <net/dst_ops.h>
14c85021	12	#include <linux/netdevice.h>
1da177e4 LT	13	#include <linux/rtnetlink.h>
	14	#include <linux/rcupdate.h>
	15	#include <linux/jiffies.h>
	16	#include <net/neighbour.h>
	17	#include <asm/processor.h>
	18
	19	/*
	20	* 0 - no debugging messages
	21	* 1 - rare events and bugs (default)
	22	* 2 - trace mode.
	23	*/
	24	#define RT_CACHE_DEBUG 0
	25
	26	#define DST_GC_MIN (HZ/10)
	27	#define DST_GC_INC (HZ/2)
	28	#define DST_GC_MAX (120*HZ)
	29
	30	/* Each dst_entry has reference count and sits in some parent list(s).
	31	* When it is removed from parent list, it is "freed" (dst_free).
	32	* After this it enters dead state (dst->obsolete > 0) and if its refcnt
	33	* is zero, it can be destroyed immediately, otherwise it is added
	34	* to gc list and garbage collector periodically checks the refcnt.
	35	*/
	36
	37	struct sk_buff;
	38
fd2c3ef7	39	struct dst_entry {
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	};
1da177e4 LT	104
1da177e4 LT	105	#ifdef __KERNEL__
	106
	107	static inline u32
	108	dst_metric(const struct dst_entry *dst, int metric)
	109	{
	110	return dst->metrics[metric-1];
	111	}
	112
0c3adfb8 GBY	113	static inline u32
	114	dst_feature(const struct dst_entry *dst, u32 feature)
	115	{
	116	return dst_metric(dst, RTAX_FEATURES) & feature;
	117	}
	118
1da177e4 LT	119	static inline u32 dst_mtu(const struct dst_entry *dst)
	120	{
	121	u32 mtu = dst_metric(dst, RTAX_MTU);
	122	/*
	123	* Alexey put it here, so ask him about it :)
	124	*/
	125	barrier();
	126	return mtu;
	127	}
	128
c1e20f7c SH	129	/* RTT metrics are stored in milliseconds for user ABI, but used as jiffies */
	130	static inline unsigned long dst_metric_rtt(const struct dst_entry *dst, int metric)
	131	{
	132	return msecs_to_jiffies(dst_metric(dst, metric));
	133	}
	134
	135	static inline void set_dst_metric_rtt(struct dst_entry *dst, int metric,
	136	unsigned long rtt)
	137	{
	138	dst->metrics[metric-1] = jiffies_to_msecs(rtt);
	139	}
	140
1da177e4 LT	141	static inline u32
	142	dst_allfrag(const struct dst_entry *dst)
	143	{
0c3adfb8	144	int ret = dst_feature(dst, RTAX_FEATURE_ALLFRAG);
1da177e4 LT	145	/* Yes, _exactly_. This is paranoia. */
	146	barrier();
	147	return ret;
	148	}
	149
	150	static inline int
	151	dst_metric_locked(struct dst_entry *dst, int metric)
	152	{
	153	return dst_metric(dst, RTAX_LOCK) & (1<<metric);
	154	}
	155
	156	static inline void dst_hold(struct dst_entry * dst)
	157	{
5635c10d ED	158	/*
	159	* If your kernel compilation stops here, please check
	160	* __pad_to_align_refcnt declaration in struct dst_entry
	161	*/
	162	BUILD_BUG_ON(offsetof(struct dst_entry, __refcnt) & 63);
1da177e4 LT	163	atomic_inc(&dst->__refcnt);
	164	}
	165
03f49f34 PE	166	static inline void dst_use(struct dst_entry *dst, unsigned long time)
	167	{
	168	dst_hold(dst);
	169	dst->__use++;
	170	dst->lastuse = time;
	171	}
	172
1da177e4 LT	173	static inline
	174	struct dst_entry * dst_clone(struct dst_entry * dst)
	175	{
	176	if (dst)
	177	atomic_inc(&dst->__refcnt);
	178	return dst;
	179	}
	180
8d330868	181	extern void dst_release(struct dst_entry *dst);
adf30907 ED	182	static inline void skb_dst_drop(struct sk_buff *skb)
	183	{
	184	if (skb->_skb_dst)
	185	dst_release(skb_dst(skb));
	186	skb->_skb_dst = 0UL;
	187	}
1da177e4 LT	188
	189	/* Children define the path of the packet through the
	190	* Linux networking. Thus, destinations are stackable.
	191	*/
	192
	193	static inline struct dst_entry dst_pop(struct dst_entry dst)
	194	{
	195	struct dst_entry *child = dst_clone(dst->child);
	196
	197	dst_release(dst);
	198	return child;
	199	}
	200
352e512c	201	extern int dst_discard(struct sk_buff *skb);
1da177e4 LT	202	extern void * dst_alloc(struct dst_ops * ops);
	203	extern void __dst_free(struct dst_entry * dst);
	204	extern struct dst_entry dst_destroy(struct dst_entry dst);
	205
	206	static inline void dst_free(struct dst_entry * dst)
	207	{
	208	if (dst->obsolete > 1)
	209	return;
	210	if (!atomic_read(&dst->__refcnt)) {
	211	dst = dst_destroy(dst);
	212	if (!dst)
	213	return;
	214	}
	215	__dst_free(dst);
	216	}
	217
	218	static inline void dst_rcu_free(struct rcu_head *head)
	219	{
	220	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
	221	dst_free(dst);
	222	}
	223
	224	static inline void dst_confirm(struct dst_entry *dst)
	225	{
	226	if (dst)
	227	neigh_confirm(dst->neighbour);
	228	}
	229
ea94ff3b KK	230	static inline void dst_negative_advice(struct dst_entry **dst_p,
ea94ff3b KK	231	struct sock *sk)
1da177e4 LT	232	{
1da177e4 LT	233	struct dst_entry * dst = *dst_p;
ea94ff3b	234	if (dst && dst->ops->negative_advice) {
1da177e4	235	*dst_p = dst->ops->negative_advice(dst);
ea94ff3b KK	236
	237	if (dst != *dst_p) {
	238	extern void sk_reset_txq(struct sock *sk);
	239
	240	sk_reset_txq(sk);
	241	}
	242	}
1da177e4 LT	243	}
	244
	245	static inline void dst_link_failure(struct sk_buff *skb)
	246	{
adf30907	247	struct dst_entry *dst = skb_dst(skb);
1da177e4 LT	248	if (dst && dst->ops && dst->ops->link_failure)
	249	dst->ops->link_failure(skb);
	250	}
	251
	252	static inline void dst_set_expires(struct dst_entry *dst, int timeout)
	253	{
	254	unsigned long expires = jiffies + timeout;
	255
	256	if (expires == 0)
	257	expires = 1;
	258
	259	if (dst->expires == 0 \|\| time_before(expires, dst->expires))
	260	dst->expires = expires;
	261	}
	262
	263	/* Output packet to network from transport. */
	264	static inline int dst_output(struct sk_buff *skb)
	265	{
adf30907	266	return skb_dst(skb)->output(skb);
1da177e4 LT	267	}
	268
	269	/* Input packet from network to transport. */
	270	static inline int dst_input(struct sk_buff *skb)
	271	{
adf30907	272	return skb_dst(skb)->input(skb);
1da177e4 LT	273	}
	274
	275	static inline struct dst_entry dst_check(struct dst_entry dst, u32 cookie)
	276	{
	277	if (dst->obsolete)
	278	dst = dst->ops->check(dst, cookie);
	279	return dst;
	280	}
	281
	282	extern void dst_init(void);
	283
815f4e57 HX	284	/* Flags for xfrm_lookup flags argument. */
	285	enum {
	286	XFRM_LOOKUP_WAIT = 1 << 0,
8b7817f3	287	XFRM_LOOKUP_ICMP = 1 << 1,
815f4e57 HX	288	};
815f4e57 HX	289
1da177e4 LT	290	struct flowi;
1da177e4 LT	291	#ifndef CONFIG_XFRM
52479b62 AD	292	static inline int xfrm_lookup(struct net net, struct dst_entry *dst_p,
52479b62 AD	293	struct flowi fl, struct sock sk, int flags)
1da177e4 LT	294	{
	295	return 0;
	296	}
52479b62 AD	297	static inline int __xfrm_lookup(struct net net, struct dst_entry *dst_p,
52479b62 AD	298	struct flowi fl, struct sock sk, int flags)
14e50e57 DM	299	{
	300	return 0;
	301	}
1da177e4	302	#else
52479b62 AD	303	extern int xfrm_lookup(struct net net, struct dst_entry *dst_p,
	304	struct flowi fl, struct sock sk, int flags);
	305	extern int __xfrm_lookup(struct net net, struct dst_entry *dst_p,
	306	struct flowi fl, struct sock sk, int flags);
1da177e4 LT	307	#endif
	308	#endif
	309
	310	#endif /* _NET_DST_H */