neighbour.c

Linux内核源代码 为压缩文件 是<<Linux内核>>一书中的源代码

static void neigh_sync(struct neighbour *n)
{
	unsigned long now = jiffies;
	u8 state = n->nud_state;

	ASSERT_WL(n);
	if (state&(NUD_NOARP|NUD_PERMANENT))
		return;
	if (state&NUD_REACHABLE) {
		if (now - n->confirmed > n->parms->reachable_time) {
			n->nud_state = NUD_STALE;
			neigh_suspect(n);
		}
	} else if (state&NUD_VALID) {
		if (now - n->confirmed < n->parms->reachable_time) {
			neigh_del_timer(n);
			n->nud_state = NUD_REACHABLE;
			neigh_connect(n);
		}
	}
}

static void SMP_TIMER_NAME(neigh_periodic_timer)(unsigned long arg)
{
	struct neigh_table *tbl = (struct neigh_table*)arg;
	unsigned long now = jiffies;
	int i;


	write_lock(&tbl->lock);

	/*
	 *	periodicly recompute ReachableTime from random function
	 */
	
	if (now - tbl->last_rand > 300*HZ) {
		struct neigh_parms *p;
		tbl->last_rand = now;
		for (p=&tbl->parms; p; p = p->next)
			p->reachable_time = neigh_rand_reach_time(p->base_reachable_time);
	}

	for (i=0; i <= NEIGH_HASHMASK; i++) {
		struct neighbour *n, **np;

		np = &tbl->hash_buckets[i];
		while ((n = *np) != NULL) {
			unsigned state;

			write_lock(&n->lock);

			state = n->nud_state;
			if (state&(NUD_PERMANENT|NUD_IN_TIMER)) {
				write_unlock(&n->lock);
				goto next_elt;
			}

			if ((long)(n->used - n->confirmed) < 0)
				n->used = n->confirmed;

			if (atomic_read(&n->refcnt) == 1 &&
			    (state == NUD_FAILED || now - n->used > n->parms->gc_staletime)) {
				*np = n->next;
				n->dead = 1;
				write_unlock(&n->lock);
				neigh_release(n);
				continue;
			}

			if (n->nud_state&NUD_REACHABLE &&
			    now - n->confirmed > n->parms->reachable_time) {
				n->nud_state = NUD_STALE;
				neigh_suspect(n);
			}
			write_unlock(&n->lock);

next_elt:
			np = &n->next;
		}
	}

	mod_timer(&tbl->gc_timer, now + tbl->gc_interval);
	write_unlock(&tbl->lock);
}

#ifdef CONFIG_SMP
static void neigh_periodic_timer(unsigned long arg)
{
	struct neigh_table *tbl = (struct neigh_table*)arg;
	
	tasklet_schedule(&tbl->gc_task);
}
#endif

static __inline__ int neigh_max_probes(struct neighbour *n)
{
	struct neigh_parms *p = n->parms;
	return p->ucast_probes + p->app_probes + p->mcast_probes;
}


/* Called when a timer expires for a neighbour entry. */

static void neigh_timer_handler(unsigned long arg) 
{
	unsigned long now = jiffies;
	struct neighbour *neigh = (struct neighbour*)arg;
	unsigned state;
	int notify = 0;

	write_lock(&neigh->lock);

	state = neigh->nud_state;

	if (!(state&NUD_IN_TIMER)) {
#ifndef CONFIG_SMP
		printk("neigh: timer & !nud_in_timer\n");
#endif
		goto out;
	}

	if ((state&NUD_VALID) &&
	    now - neigh->confirmed < neigh->parms->reachable_time) {
		neigh->nud_state = NUD_REACHABLE;
		NEIGH_PRINTK2("neigh %p is still alive.\n", neigh);
		neigh_connect(neigh);
		goto out;
	}
	if (state == NUD_DELAY) {
		NEIGH_PRINTK2("neigh %p is probed.\n", neigh);
		neigh->nud_state = NUD_PROBE;
		atomic_set(&neigh->probes, 0);
	}

	if (atomic_read(&neigh->probes) >= neigh_max_probes(neigh)) {
		struct sk_buff *skb;

		neigh->nud_state = NUD_FAILED;
		notify = 1;
		neigh->tbl->stats.res_failed++;
		NEIGH_PRINTK2("neigh %p is failed.\n", neigh);

		/* It is very thin place. report_unreachable is very complicated
		   routine. Particularly, it can hit the same neighbour entry!
		   
		   So that, we try to be accurate and avoid dead loop. --ANK
		 */
		while(neigh->nud_state==NUD_FAILED && (skb=__skb_dequeue(&neigh->arp_queue)) != NULL) {
			write_unlock(&neigh->lock);
			neigh->ops->error_report(neigh, skb);
			write_lock(&neigh->lock);
		}
		skb_queue_purge(&neigh->arp_queue);
		goto out;
	}

	neigh->timer.expires = now + neigh->parms->retrans_time;
	add_timer(&neigh->timer);
	write_unlock(&neigh->lock);

	neigh->ops->solicit(neigh, skb_peek(&neigh->arp_queue));
	atomic_inc(&neigh->probes);
	return;

out:
	write_unlock(&neigh->lock);
#ifdef CONFIG_ARPD
	if (notify && neigh->parms->app_probes)
		neigh_app_notify(neigh);
#endif
	neigh_release(neigh);
}

int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb)
{
	write_lock_bh(&neigh->lock);
	if (!(neigh->nud_state&(NUD_CONNECTED|NUD_DELAY|NUD_PROBE))) {
		if (!(neigh->nud_state&(NUD_STALE|NUD_INCOMPLETE))) {
			if (neigh->parms->mcast_probes + neigh->parms->app_probes) {
				atomic_set(&neigh->probes, neigh->parms->ucast_probes);
				neigh->nud_state = NUD_INCOMPLETE;
				neigh_hold(neigh);
				neigh->timer.expires = jiffies + neigh->parms->retrans_time;
				add_timer(&neigh->timer);
				write_unlock_bh(&neigh->lock);
				neigh->ops->solicit(neigh, skb);
				atomic_inc(&neigh->probes);
				write_lock_bh(&neigh->lock);
			} else {
				neigh->nud_state = NUD_FAILED;
				write_unlock_bh(&neigh->lock);

				if (skb)
					kfree_skb(skb);
				return 1;
			}
		}
		if (neigh->nud_state == NUD_INCOMPLETE) {
			if (skb) {
				if (skb_queue_len(&neigh->arp_queue) >= neigh->parms->queue_len) {
					struct sk_buff *buff;
					buff = neigh->arp_queue.prev;
					__skb_unlink(buff, &neigh->arp_queue);
					kfree_skb(buff);
				}
				__skb_queue_head(&neigh->arp_queue, skb);
			}
			write_unlock_bh(&neigh->lock);
			return 1;
		}
		if (neigh->nud_state == NUD_STALE) {
			NEIGH_PRINTK2("neigh %p is delayed.\n", neigh);
			neigh_hold(neigh);
			neigh->nud_state = NUD_DELAY;
			neigh->timer.expires = jiffies + neigh->parms->delay_probe_time;
			add_timer(&neigh->timer);
		}
	}
	write_unlock_bh(&neigh->lock);
	return 0;
}

static __inline__ void neigh_update_hhs(struct neighbour *neigh)
{
	struct hh_cache *hh;
	void (*update)(struct hh_cache*, struct net_device*, unsigned char*) =
		neigh->dev->header_cache_update;

	if (update) {
		for (hh=neigh->hh; hh; hh=hh->hh_next) {
			write_lock_bh(&hh->hh_lock);
			update(hh, neigh->dev, neigh->ha);
			write_unlock_bh(&hh->hh_lock);
		}
	}
}



/* Generic update routine.
   -- lladdr is new lladdr or NULL, if it is not supplied.
   -- new    is new state.
   -- override==1 allows to override existing lladdr, if it is different.
   -- arp==0 means that the change is administrative.

   Caller MUST hold reference count on the entry.
 */

int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, int override, int arp)
{
	u8 old;
	int err;
	int notify = 0;
	struct net_device *dev = neigh->dev;

	write_lock_bh(&neigh->lock);
	old = neigh->nud_state;

	err = -EPERM;
	if (arp && (old&(NUD_NOARP|NUD_PERMANENT)))
		goto out;

	if (!(new&NUD_VALID)) {
		neigh_del_timer(neigh);
		if (old&NUD_CONNECTED)
			neigh_suspect(neigh);
		neigh->nud_state = new;
		err = 0;
		notify = old&NUD_VALID;
		goto out;
	}

	/* Compare new lladdr with cached one */
	if (dev->addr_len == 0) {
		/* First case: device needs no address. */
		lladdr = neigh->ha;
	} else if (lladdr) {
		/* The second case: if something is already cached
		   and a new address is proposed:
		   - compare new & old
		   - if they are different, check override flag
		 */
		if (old&NUD_VALID) {
			if (memcmp(lladdr, neigh->ha, dev->addr_len) == 0)
				lladdr = neigh->ha;
			else if (!override)
				goto out;
		}
	} else {
		/* No address is supplied; if we know something,
		   use it, otherwise discard the request.
		 */
		err = -EINVAL;
		if (!(old&NUD_VALID))
			goto out;
		lladdr = neigh->ha;
	}

	neigh_sync(neigh);
	old = neigh->nud_state;
	if (new&NUD_CONNECTED)
		neigh->confirmed = jiffies;
	neigh->updated = jiffies;

	/* If entry was valid and address is not changed,
	   do not change entry state, if new one is STALE.
	 */
	err = 0;
	if (old&NUD_VALID) {
		if (lladdr == neigh->ha)
			if (new == old || (new == NUD_STALE && (old&NUD_CONNECTED)))
				goto out;
	}
	neigh_del_timer(neigh);
	neigh->nud_state = new;
	if (lladdr != neigh->ha) {
		memcpy(&neigh->ha, lladdr, dev->addr_len);
		neigh_update_hhs(neigh);
		if (!(new&NUD_CONNECTED))
			neigh->confirmed = jiffies - (neigh->parms->base_reachable_time<<1);
#ifdef CONFIG_ARPD
		notify = 1;
#endif
	}
	if (new == old)
		goto out;
	if (new&NUD_CONNECTED)
		neigh_connect(neigh);
	else
		neigh_suspect(neigh);
	if (!(old&NUD_VALID)) {
		struct sk_buff *skb;

		/* Again: avoid dead loop if something went wrong */

		while (neigh->nud_state&NUD_VALID &&
		       (skb=__skb_dequeue(&neigh->arp_queue)) != NULL) {
			struct neighbour *n1 = neigh;
			write_unlock_bh(&neigh->lock);
			/* On shaper/eql skb->dst->neighbour != neigh :( */
			if (skb->dst && skb->dst->neighbour)
				n1 = skb->dst->neighbour;
			n1->output(skb);
			write_lock_bh(&neigh->lock);
		}
		skb_queue_purge(&neigh->arp_queue);
	}
out:
	write_unlock_bh(&neigh->lock);
#ifdef CONFIG_ARPD
	if (notify && neigh->parms->app_probes)
		neigh_app_notify(neigh);
#endif
	return err;
}

struct neighbour * neigh_event_ns(struct neigh_table *tbl,
				  u8 *lladdr, void *saddr,
				  struct net_device *dev)
{
	struct neighbour *neigh;

	neigh = __neigh_lookup(tbl, saddr, dev, lladdr || !dev->addr_len);
	if (neigh)
		neigh_update(neigh, lladdr, NUD_STALE, 1, 1);
	return neigh;
}

static void neigh_hh_init(struct neighbour *n, struct dst_entry *dst, u16 protocol)
{
	struct hh_cache	*hh = NULL;
	struct net_device *dev = dst->dev;

	for (hh=n->hh; hh; hh = hh->hh_next)
		if (hh->hh_type == protocol)
			break;

	if (!hh && (hh = kmalloc(sizeof(*hh), GFP_ATOMIC)) != NULL) {
		memset(hh, 0, sizeof(struct hh_cache));
		hh->hh_lock = RW_LOCK_UNLOCKED;
		hh->hh_type = protocol;
		atomic_set(&hh->hh_refcnt, 0);
		hh->hh_next = NULL;
		if (dev->hard_header_cache(n, hh)) {
			kfree(hh);
			hh = NULL;
		} else {
			atomic_inc(&hh->hh_refcnt);
			hh->hh_next = n->hh;
			n->hh = hh;
			if (n->nud_state&NUD_CONNECTED)
				hh->hh_output = n->ops->hh_output;
			else
				hh->hh_output = n->ops->output;
		}
	}
	if (hh)	{
		atomic_inc(&hh->hh_refcnt);
		dst->hh = hh;
	}
}

/* This function can be used in contexts, where only old dev_queue_xmit
   worked, f.e. if you want to override normal output path (eql, shaper),
   but resoltution is not made yet.
 */

int neigh_compat_output(struct sk_buff *skb)
{
	struct net_device *dev = skb->dev;

	__skb_pull(skb, skb->nh.raw - skb->data);

	if (dev->hard_header &&
	    dev->hard_header(skb, dev, ntohs(skb->protocol), NULL, NULL, skb->len) < 0 &&
	    dev->rebuild_header(skb))
		return 0;

	return dev_queue_xmit(skb);
}

/* Slow and careful. */

int neigh_resolve_output(struct sk_buff *skb)
{
	struct dst_entry *dst = skb->dst;
	struct neighbour *neigh;

	if (!dst || !(neigh = dst->neighbour))
		goto discard;

	__skb_pull(skb, skb->nh.raw - skb->data);

	if (neigh_event_send(neigh, skb) == 0) {
		int err;
		struct net_device *dev = neigh->dev;
		if (dev->hard_header_cache && dst->hh == NULL) {
			write_lock_bh(&neigh->lock);
			if (dst->hh == NULL)
				neigh_hh_init(neigh, dst, dst->ops->protocol);
			err = dev->hard_header(skb, dev, ntohs(skb->protocol), neigh->ha, NULL, skb->len);
			write_unlock_bh(&neigh->lock);
		} else {
			read_lock_bh(&neigh->lock);
			err = dev->hard_header(skb, dev, ntohs(skb->protocol), neigh->ha, NULL, skb->len);
			read_unlock_bh(&neigh->lock);
		}
		if (err >= 0)
			return neigh->ops->queue_xmit(skb);
		kfree_skb(skb);
		return -EINVAL;
	}
	return 0;

discard:
	NEIGH_PRINTK1("neigh_resolve_output: dst=%p neigh=%p\n", dst, dst ? dst->neighbour : NULL);
	kfree_skb(skb);
	return -EINVAL;
}

/* As fast as possible without hh cache */

int neigh_connected_output(struct sk_buff *skb)
{
	int err;
	struct dst_entry *dst = skb->dst;
	struct neighbour *neigh = dst->neighbour;
	struct net_device *dev = neigh->dev;

	__skb_pull(skb, skb->nh.raw - skb->data);

	read_lock_bh(&neigh->lock);
	err = dev->hard_header(skb, dev, ntohs(skb->protocol), neigh->ha, NULL, skb->len);
	read_unlock_bh(&neigh->lock);
	if (err >= 0)
		return neigh->ops->queue_xmit(skb);
	kfree_skb(skb);
	return -EINVAL;
}

static void neigh_proxy_process(unsigned long arg)
{
	struct neigh_table *tbl = (struct neigh_table *)arg;
	long sched_next = 0;
	unsigned long now = jiffies;
	struct sk_buff *skb;

	spin_lock(&tbl->proxy_queue.lock);

	skb = tbl->proxy_queue.next;

	while (skb != (struct sk_buff*)&tbl->proxy_queue) {
		struct sk_buff *back = skb;
		long tdif = back->stamp.tv_usec - now;

		skb = skb->next;
		if (tdif <= 0) {
			struct net_device *dev = back->dev;
			__skb_unlink(back, &tbl->proxy_queue);
			if (tbl->proxy_redo && netif_running(dev))
				tbl->proxy_redo(back);
			else
				kfree_skb(back);

			dev_put(dev);
		} else if (!sched_next || tdif < sched_next)
			sched_next = tdif;
	}
	del_timer(&tbl->proxy_timer);
	if (sched_next)
		mod_timer(&tbl->proxy_timer, jiffies + sched_next);
	spin_unlock(&tbl->proxy_queue.lock);
}

void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p,
		    struct sk_buff *skb)
{
	unsigned long now = jiffies;
	long sched_next = net_random()%p->proxy_delay;

	if (tbl->proxy_queue.qlen > p->proxy_qlen) {
		kfree_skb(skb);
		return;
	}
	skb->stamp.tv_sec = 0;
	skb->stamp.tv_usec = now + sched_next;