[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 <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		0x0001
#define DST_NOXFRM		0x0002
#define DST_NOPOLICY		0x0004
#define DST_NOHASH		0x0008
#define DST_NOCACHE		0x0010
	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[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 __rcu	*rt_next;
		struct rt6_info		*rt6_next;
		struct dn_route __rcu	*dn_next;
	};
};

#ifdef __KERNEL__

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

static inline void dst_metric_set(struct dst_entry *dst, int metric, u32 val)
{
	dst->_metrics[metric-1] = val;
}

static inline void dst_import_metrics(struct dst_entry *dst, const u32 *src_metrics)
{
	memcpy(dst->_metrics, src_metrics, RTAX_MAX * sizeof(u32));
}

static inline void dst_copy_metrics(struct dst_entry *dest, const struct dst_entry *src)
{
	dst_import_metrics(dest, src->_metrics);
}

static inline u32 *dst_metrics_ptr(struct dst_entry *dst)
{
	return dst->_metrics;
}

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_metric_set(dst, metric, 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 void dst_use_noref(struct dst_entry *dst, unsigned long time)
{
	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 refdst_drop(unsigned long refdst)
{
	if (!(refdst & SKB_DST_NOREF))
		dst_release((struct dst_entry *)(refdst & SKB_DST_PTRMASK));
}

/**
 * skb_dst_drop - drops skb dst
 * @skb: buffer
 *
 * Drops dst reference count if a reference was taken.
 */
static inline void skb_dst_drop(struct sk_buff *skb)
{
	if (skb->_skb_refdst) {
		refdst_drop(skb->_skb_refdst);
		skb->_skb_refdst = 0UL;
	}
}

static inline void skb_dst_copy(struct sk_buff *nskb, const struct sk_buff *oskb)
{
	nskb->_skb_refdst = oskb->_skb_refdst;
	if (!(nskb->_skb_refdst & SKB_DST_NOREF))
		dst_clone(skb_dst(nskb));
}

/**
 * skb_dst_force - makes sure skb dst is refcounted
 * @skb: buffer
 *
 * If dst is not yet refcounted, let's do it
 */
static inline void skb_dst_force(struct sk_buff *skb)
{
	if (skb_dst_is_noref(skb)) {
		WARN_ON(!rcu_read_lock_held());
		skb->_skb_refdst &= ~SKB_DST_NOREF;
		dst_clone(skb_dst(skb));
	}
}


/**
 *	__skb_tunnel_rx - prepare skb for rx reinsert
 *	@skb: buffer
 *	@dev: tunnel device
 *
 *	After decapsulation, packet is going to re-enter (netif_rx()) our stack,
 *	so make some cleanups. (no accounting done)
 */
static inline void __skb_tunnel_rx(struct sk_buff *skb, struct net_device *dev)
{
	skb->dev = dev;
	skb->rxhash = 0;
	skb_set_queue_mapping(skb, 0);
	skb_dst_drop(skb);
	nf_reset(skb);
}

/**
 *	skb_tunnel_rx - prepare skb for rx reinsert
 *	@skb: buffer
 *	@dev: tunnel device
 *
 *	After decapsulation, packet is going to re-enter (netif_rx()) our stack,
 *	so make some cleanups, and perform accounting.
 *	Note: this accounting is not SMP safe.
 */
static inline void skb_tunnel_rx(struct sk_buff *skb, struct net_device *dev)
{
	/* TODO : stats should be SMP safe */
	dev->stats.rx_packets++;
	dev->stats.rx_bytes += skb->len;
	__skb_tunnel_rx(skb, dev);
}

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

static inline struct dst_entry *skb_dst_pop(struct sk_buff *skb)
{
	struct dst_entry *child = skb_dst(skb)->child;

	skb_dst_drop(skb);
	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_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
	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
	11	#include <net/dst_ops.h>
	12	#include <linux/netdevice.h>
	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
	39	struct dst_entry {
	40	struct rcu_head rcu_head;
	41	struct dst_entry *child;
	42	struct net_device *dev;
	43	short error;
	44	short obsolete;
	45	int flags;
	46	#define DST_HOST 0x0001
	47	#define DST_NOXFRM 0x0002
	48	#define DST_NOPOLICY 0x0004
	49	#define DST_NOHASH 0x0008
	50	#define DST_NOCACHE 0x0010
	51	unsigned long expires;
	52
	53	unsigned short header_len; /* more space at head required */
	54	unsigned short trailer_len; /* space to reserve at tail */
	55
	56	unsigned int rate_tokens;
	57	unsigned long rate_last; /* rate limiting for ICMP */
	58
	59	struct dst_entry *path;
	60
	61	struct neighbour *neighbour;
	62	struct hh_cache *hh;
	63	#ifdef CONFIG_XFRM
	64	struct xfrm_state *xfrm;
	65	#else
	66	void *__pad1;
	67	#endif
	68	int (input)(struct sk_buff);
	69	int (output)(struct sk_buff);
	70
	71	struct dst_ops *ops;
	72
	73	u32 _metrics[RTAX_MAX];
	74
	75	#ifdef CONFIG_NET_CLS_ROUTE
	76	__u32 tclassid;
	77	#else
	78	__u32 __pad2;
	79	#endif
	80
	81
	82	/*
	83	* Align __refcnt to a 64 bytes alignment
	84	* (L1_CACHE_SIZE would be too much)
	85	*/
	86	#ifdef CONFIG_64BIT
	87	long __pad_to_align_refcnt[1];
	88	#endif
	89	/*
	90	* __refcnt wants to be on a different cache line from
	91	* input/output/ops or performance tanks badly
	92	*/
	93	atomic_t __refcnt; /* client references */
	94	int __use;
	95	unsigned long lastuse;
	96	union {
	97	struct dst_entry *next;
	98	struct rtable __rcu *rt_next;
	99	struct rt6_info *rt6_next;
	100	struct dn_route __rcu *dn_next;
	101	};
	102	};
	103
	104	#ifdef __KERNEL__
	105
	106	static inline u32
	107	dst_metric(const struct dst_entry *dst, int metric)
	108	{
	109	return dst->_metrics[metric-1];
	110	}
	111
	112	static inline void dst_metric_set(struct dst_entry *dst, int metric, u32 val)
	113	{
	114	dst->_metrics[metric-1] = val;
	115	}
	116
	117	static inline void dst_import_metrics(struct dst_entry dst, const u32 src_metrics)
	118	{
	119	memcpy(dst->_metrics, src_metrics, RTAX_MAX * sizeof(u32));
	120	}
	121
	122	static inline void dst_copy_metrics(struct dst_entry dest, const struct dst_entry src)
	123	{
	124	dst_import_metrics(dest, src->_metrics);
	125	}
	126
	127	static inline u32 dst_metrics_ptr(struct dst_entry dst)
	128	{
	129	return dst->_metrics;
	130	}
	131
	132	static inline u32
	133	dst_feature(const struct dst_entry *dst, u32 feature)
	134	{
	135	return dst_metric(dst, RTAX_FEATURES) & feature;
	136	}
	137
	138	static inline u32 dst_mtu(const struct dst_entry *dst)
	139	{
	140	u32 mtu = dst_metric(dst, RTAX_MTU);
	141	/*
	142	* Alexey put it here, so ask him about it :)
	143	*/
	144	barrier();
	145	return mtu;
	146	}
	147
	148	/* RTT metrics are stored in milliseconds for user ABI, but used as jiffies */
	149	static inline unsigned long dst_metric_rtt(const struct dst_entry *dst, int metric)
	150	{
	151	return msecs_to_jiffies(dst_metric(dst, metric));
	152	}
	153
	154	static inline void set_dst_metric_rtt(struct dst_entry *dst, int metric,
	155	unsigned long rtt)
	156	{
	157	dst_metric_set(dst, metric, jiffies_to_msecs(rtt));
	158	}
	159
	160	static inline u32
	161	dst_allfrag(const struct dst_entry *dst)
	162	{
	163	int ret = dst_feature(dst, RTAX_FEATURE_ALLFRAG);
	164	/* Yes, _exactly_. This is paranoia. */
	165	barrier();
	166	return ret;
	167	}
	168
	169	static inline int
	170	dst_metric_locked(struct dst_entry *dst, int metric)
	171	{
	172	return dst_metric(dst, RTAX_LOCK) & (1<<metric);
	173	}
	174
	175	static inline void dst_hold(struct dst_entry * dst)
	176	{
	177	/*
	178	* If your kernel compilation stops here, please check
	179	* __pad_to_align_refcnt declaration in struct dst_entry
	180	*/
	181	BUILD_BUG_ON(offsetof(struct dst_entry, __refcnt) & 63);
	182	atomic_inc(&dst->__refcnt);
	183	}
	184
	185	static inline void dst_use(struct dst_entry *dst, unsigned long time)
	186	{
	187	dst_hold(dst);
	188	dst->__use++;
	189	dst->lastuse = time;
	190	}
	191
	192	static inline void dst_use_noref(struct dst_entry *dst, unsigned long time)
	193	{
	194	dst->__use++;
	195	dst->lastuse = time;
	196	}
	197
	198	static inline
	199	struct dst_entry * dst_clone(struct dst_entry * dst)
	200	{
	201	if (dst)
	202	atomic_inc(&dst->__refcnt);
	203	return dst;
	204	}
	205
	206	extern void dst_release(struct dst_entry *dst);
	207
	208	static inline void refdst_drop(unsigned long refdst)
	209	{
	210	if (!(refdst & SKB_DST_NOREF))
	211	dst_release((struct dst_entry *)(refdst & SKB_DST_PTRMASK));
	212	}
	213
	214	/**
	215	* skb_dst_drop - drops skb dst
	216	* @skb: buffer
	217	*
	218	* Drops dst reference count if a reference was taken.
	219	*/
	220	static inline void skb_dst_drop(struct sk_buff *skb)
	221	{
	222	if (skb->_skb_refdst) {
	223	refdst_drop(skb->_skb_refdst);
	224	skb->_skb_refdst = 0UL;
	225	}
	226	}
	227
	228	static inline void skb_dst_copy(struct sk_buff nskb, const struct sk_buff oskb)
	229	{
	230	nskb->_skb_refdst = oskb->_skb_refdst;
	231	if (!(nskb->_skb_refdst & SKB_DST_NOREF))
	232	dst_clone(skb_dst(nskb));
	233	}
	234
	235	/**
	236	* skb_dst_force - makes sure skb dst is refcounted
	237	* @skb: buffer
	238	*
	239	* If dst is not yet refcounted, let's do it
	240	*/
	241	static inline void skb_dst_force(struct sk_buff *skb)
	242	{
	243	if (skb_dst_is_noref(skb)) {
	244	WARN_ON(!rcu_read_lock_held());
	245	skb->_skb_refdst &= ~SKB_DST_NOREF;
	246	dst_clone(skb_dst(skb));
	247	}
	248	}
	249
	250
	251	/**
	252	* __skb_tunnel_rx - prepare skb for rx reinsert
	253	* @skb: buffer
	254	* @dev: tunnel device
	255	*
	256	* After decapsulation, packet is going to re-enter (netif_rx()) our stack,
	257	* so make some cleanups. (no accounting done)
	258	*/
	259	static inline void __skb_tunnel_rx(struct sk_buff skb, struct net_device dev)
	260	{
	261	skb->dev = dev;
	262	skb->rxhash = 0;
	263	skb_set_queue_mapping(skb, 0);
	264	skb_dst_drop(skb);
	265	nf_reset(skb);
	266	}
	267
	268	/**
	269	* skb_tunnel_rx - prepare skb for rx reinsert
	270	* @skb: buffer
	271	* @dev: tunnel device
	272	*
	273	* After decapsulation, packet is going to re-enter (netif_rx()) our stack,
	274	* so make some cleanups, and perform accounting.
	275	* Note: this accounting is not SMP safe.
	276	*/
	277	static inline void skb_tunnel_rx(struct sk_buff skb, struct net_device dev)
	278	{
	279	/* TODO : stats should be SMP safe */
	280	dev->stats.rx_packets++;
	281	dev->stats.rx_bytes += skb->len;
	282	__skb_tunnel_rx(skb, dev);
	283	}
	284
	285	/* Children define the path of the packet through the
	286	* Linux networking. Thus, destinations are stackable.
	287	*/
	288
	289	static inline struct dst_entry skb_dst_pop(struct sk_buff skb)
	290	{
	291	struct dst_entry *child = skb_dst(skb)->child;
	292
	293	skb_dst_drop(skb);
	294	return child;
	295	}
	296
	297	extern int dst_discard(struct sk_buff *skb);
	298	extern void * dst_alloc(struct dst_ops * ops);
	299	extern void __dst_free(struct dst_entry * dst);
	300	extern struct dst_entry dst_destroy(struct dst_entry dst);
	301
	302	static inline void dst_free(struct dst_entry * dst)
	303	{
	304	if (dst->obsolete > 1)
	305	return;
	306	if (!atomic_read(&dst->__refcnt)) {
	307	dst = dst_destroy(dst);
	308	if (!dst)
	309	return;
	310	}
	311	__dst_free(dst);
	312	}
	313
	314	static inline void dst_rcu_free(struct rcu_head *head)
	315	{
	316	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
	317	dst_free(dst);
	318	}
	319
	320	static inline void dst_confirm(struct dst_entry *dst)
	321	{
	322	if (dst)
	323	neigh_confirm(dst->neighbour);
	324	}
	325
	326	static inline void dst_link_failure(struct sk_buff *skb)
	327	{
	328	struct dst_entry *dst = skb_dst(skb);
	329	if (dst && dst->ops && dst->ops->link_failure)
	330	dst->ops->link_failure(skb);
	331	}
	332
	333	static inline void dst_set_expires(struct dst_entry *dst, int timeout)
	334	{
	335	unsigned long expires = jiffies + timeout;
	336
	337	if (expires == 0)
	338	expires = 1;
	339
	340	if (dst->expires == 0 \|\| time_before(expires, dst->expires))
	341	dst->expires = expires;
	342	}
	343
	344	/* Output packet to network from transport. */
	345	static inline int dst_output(struct sk_buff *skb)
	346	{
	347	return skb_dst(skb)->output(skb);
	348	}
	349
	350	/* Input packet from network to transport. */
	351	static inline int dst_input(struct sk_buff *skb)
	352	{
	353	return skb_dst(skb)->input(skb);
	354	}
	355
	356	static inline struct dst_entry dst_check(struct dst_entry dst, u32 cookie)
	357	{
	358	if (dst->obsolete)
	359	dst = dst->ops->check(dst, cookie);
	360	return dst;
	361	}
	362
	363	extern void dst_init(void);
	364
	365	/* Flags for xfrm_lookup flags argument. */
	366	enum {
	367	XFRM_LOOKUP_WAIT = 1 << 0,
	368	XFRM_LOOKUP_ICMP = 1 << 1,
	369	};
	370
	371	struct flowi;
	372	#ifndef CONFIG_XFRM
	373	static inline int xfrm_lookup(struct net net, struct dst_entry *dst_p,
	374	struct flowi fl, struct sock sk, int flags)
	375	{
	376	return 0;
	377	}
	378	static inline int __xfrm_lookup(struct net net, struct dst_entry *dst_p,
	379	struct flowi fl, struct sock sk, int flags)
	380	{
	381	return 0;
	382	}
	383	#else
	384	extern int xfrm_lookup(struct net net, struct dst_entry *dst_p,
	385	struct flowi fl, struct sock sk, int flags);
	386	extern int __xfrm_lookup(struct net net, struct dst_entry *dst_p,
	387	struct flowi fl, struct sock sk, int flags);
	388	#endif
	389	#endif
	390
	391	#endif /* _NET_DST_H */