route.c

Linux内核源代码 为压缩文件 是<<Linux内核>>一书中的源代码

	   route or unicast forwarding path.
	 */
	if (rt->rt_type == RTN_UNICAST || rt->key.iif == 0) {
		int err = arp_bind_neighbour(&rt->u.dst);
		if (err) {
			write_unlock_bh(&rt_hash_table[hash].lock);

			if (err != -ENOBUFS) {
				rt_drop(rt);
				return err;
			}

			/* Neighbour tables are full and nothing
			   can be released. Try to shrink route cache,
			   it is most likely it holds some neighbour records.
			 */
			if (attempts-- > 0) {
				int saved_elasticity = ip_rt_gc_elasticity;
				int saved_int = ip_rt_gc_min_interval;
				ip_rt_gc_elasticity = 1;
				ip_rt_gc_min_interval = 0;
				rt_garbage_collect();
				ip_rt_gc_min_interval = saved_int;
				ip_rt_gc_elasticity = saved_elasticity;
				goto restart;
			}

			if (net_ratelimit())
				printk("Neighbour table overflow.\n");
			rt_drop(rt);
			return -ENOBUFS;
		}
	}

	rt->u.rt_next = rt_hash_table[hash].chain;
#if RT_CACHE_DEBUG >= 2
	if (rt->u.rt_next) {
		struct rtable * trt;
		printk("rt_cache @%02x: %u.%u.%u.%u", hash, NIPQUAD(rt->rt_dst));
		for (trt=rt->u.rt_next; trt; trt=trt->u.rt_next)
			printk(" . %u.%u.%u.%u", NIPQUAD(trt->rt_dst));
		printk("\n");
	}
#endif
	rt_hash_table[hash].chain = rt;
	write_unlock_bh(&rt_hash_table[hash].lock);
	*rp = rt;
	return 0;
}

void rt_bind_peer(struct rtable *rt, int create)
{
	static spinlock_t rt_peer_lock = SPIN_LOCK_UNLOCKED;
	struct inet_peer *peer;

	peer = inet_getpeer(rt->rt_dst, create);

	spin_lock_bh(&rt_peer_lock);
	if (rt->peer == NULL) {
		rt->peer = peer;
		peer = NULL;
	}
	spin_unlock_bh(&rt_peer_lock);
	if (peer)
		inet_putpeer(peer);
}

/*
 * Peer allocation may fail only in serious out-of-memory conditions.  However
 * we still can generate some output.
 * Random ID selection looks a bit dangerous because we have no chances to
 * select ID being unique in a reasonable period of time.
 * But broken packet identifier may be better than no packet at all.
 */
static void ip_select_fb_ident(struct iphdr *iph)
{
	static spinlock_t ip_fb_id_lock = SPIN_LOCK_UNLOCKED;
	static u32 ip_fallback_id;
	u32 salt;

	spin_lock_bh(&ip_fb_id_lock);
	salt = secure_ip_id(ip_fallback_id ^ iph->daddr);
	iph->id = salt & 0xFFFF;
	ip_fallback_id = salt;
	spin_unlock_bh(&ip_fb_id_lock);
}

void __ip_select_ident(struct iphdr *iph, struct dst_entry *dst)
{
	struct rtable *rt = (struct rtable *) dst;

	if (rt) {
		if (rt->peer == NULL)
			rt_bind_peer(rt, 1);

		/* If peer is attached to destination, it is never detached,
		   so that we need not to grab a lock to dereference it.
		 */
		if (rt->peer) {
			iph->id = inet_getid(rt->peer);
			return;
		}
	} else {
		printk(KERN_DEBUG "rt_bind_peer(0) @%p\n", NET_CALLER(iph));
	}

	ip_select_fb_ident(iph);
}

static void rt_del(unsigned hash, struct rtable *rt)
{
	struct rtable **rthp;

	write_lock_bh(&rt_hash_table[hash].lock);
	ip_rt_put(rt);
	for (rthp = &rt_hash_table[hash].chain; *rthp; rthp = &(*rthp)->u.rt_next) {
		if (*rthp == rt) {
			*rthp = rt->u.rt_next;
			rt_free(rt);
			break;
		}
	}
	write_unlock_bh(&rt_hash_table[hash].lock);
}

void ip_rt_redirect(u32 old_gw, u32 daddr, u32 new_gw,
		    u32 saddr, u8 tos, struct net_device *dev)
{
	int i, k;
	struct in_device *in_dev = in_dev_get(dev);
	struct rtable *rth, **rthp;
	u32  skeys[2] = { saddr, 0 };
	int  ikeys[2] = { dev->ifindex, 0 };

	tos &= IPTOS_RT_MASK;

	if (!in_dev)
		return;

	if (new_gw == old_gw || !IN_DEV_RX_REDIRECTS(in_dev)
	    || MULTICAST(new_gw) || BADCLASS(new_gw) || ZERONET(new_gw))
		goto reject_redirect;

	if (!IN_DEV_SHARED_MEDIA(in_dev)) {
		if (!inet_addr_onlink(in_dev, new_gw, old_gw))
			goto reject_redirect;
		if (IN_DEV_SEC_REDIRECTS(in_dev) && ip_fib_check_default(new_gw, dev))
			goto reject_redirect;
	} else {
		if (inet_addr_type(new_gw) != RTN_UNICAST)
			goto reject_redirect;
	}

	for (i=0; i<2; i++) {
		for (k=0; k<2; k++) {
			unsigned hash = rt_hash_code(daddr, skeys[i]^(ikeys[k]<<5), tos);

			rthp=&rt_hash_table[hash].chain;

			read_lock(&rt_hash_table[hash].lock);
			while ( (rth = *rthp) != NULL) {
				struct rtable *rt;

				if (rth->key.dst != daddr ||
				    rth->key.src != skeys[i] ||
				    rth->key.tos != tos ||
				    rth->key.oif != ikeys[k] ||
				    rth->key.iif != 0) {
					rthp = &rth->u.rt_next;
					continue;
				}

				if (rth->rt_dst != daddr ||
				    rth->rt_src != saddr ||
				    rth->u.dst.error ||
				    rth->rt_gateway != old_gw ||
				    rth->u.dst.dev != dev)
					break;

				dst_clone(&rth->u.dst);
				read_unlock(&rt_hash_table[hash].lock);

				rt = dst_alloc(&ipv4_dst_ops);
				if (rt == NULL) {
					ip_rt_put(rth);
					in_dev_put(in_dev);
					return;
				}

				/*
				 * Copy all the information.
				 */
				*rt = *rth;
				rt->u.dst.__use = 1;
				atomic_set(&rt->u.dst.__refcnt, 1);
				if (rt->u.dst.dev)
					dev_hold(rt->u.dst.dev);
				rt->u.dst.lastuse = jiffies;
				rt->u.dst.neighbour = NULL;
				rt->u.dst.hh = NULL;
				rt->u.dst.obsolete = 0;

				rt->rt_flags |= RTCF_REDIRECTED;

				/* Gateway is different ... */
				rt->rt_gateway = new_gw;

				/* Redirect received -> path was valid */
				dst_confirm(&rth->u.dst);

				if (rt->peer)
					atomic_inc(&rt->peer->refcnt);

				if (arp_bind_neighbour(&rt->u.dst) ||
				    !(rt->u.dst.neighbour->nud_state&NUD_VALID)) {
					if (rt->u.dst.neighbour)
						neigh_event_send(rt->u.dst.neighbour, NULL);
					ip_rt_put(rth);
					rt_drop(rt);
					goto do_next;
				}

				rt_del(hash, rth);
				if (!rt_intern_hash(hash, rt, &rt))
					ip_rt_put(rt);
				goto do_next;
			}
			read_unlock(&rt_hash_table[hash].lock);
		do_next:
			;
		}
	}
	in_dev_put(in_dev);
	return;

reject_redirect:
#ifdef CONFIG_IP_ROUTE_VERBOSE
	if (IN_DEV_LOG_MARTIANS(in_dev) && net_ratelimit())
		printk(KERN_INFO "Redirect from %u.%u.%u.%u on %s about %u.%u.%u.%u ignored.\n"
		       "  Advised path = %u.%u.%u.%u -> %u.%u.%u.%u, tos %02x\n",
		       NIPQUAD(old_gw), dev->name, NIPQUAD(new_gw),
		       NIPQUAD(saddr), NIPQUAD(daddr), tos);
#endif
	in_dev_put(in_dev);
}

static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst)
{
	struct rtable *rt = (struct rtable*)dst;

	if (rt != NULL) {
		if (dst->obsolete) {
			ip_rt_put(rt);
			return NULL;
		}
		if ((rt->rt_flags&RTCF_REDIRECTED) || rt->u.dst.expires) {
			unsigned hash = rt_hash_code(rt->key.dst, rt->key.src^(rt->key.oif<<5), rt->key.tos);
#if RT_CACHE_DEBUG >= 1
			printk(KERN_DEBUG "ip_rt_advice: redirect to %u.%u.%u.%u/%02x dropped\n",
				NIPQUAD(rt->rt_dst), rt->key.tos);
#endif
			rt_del(hash, rt);
			return NULL;
		}
	}
	return dst;
}

/*
 * Algorithm:
 *	1. The first ip_rt_redirect_number redirects are sent
 *	   with exponential backoff, then we stop sending them at all,
 *	   assuming that the host ignores our redirects.
 *	2. If we did not see packets requiring redirects
 *	   during ip_rt_redirect_silence, we assume that the host
 *	   forgot redirected route and start to send redirects again.
 *
 * This algorithm is much cheaper and more intelligent than dumb load limiting
 * in icmp.c.
 *
 * NOTE. Do not forget to inhibit load limiting for redirects (redundant)
 * and "frag. need" (breaks PMTU discovery) in icmp.c.
 */

void ip_rt_send_redirect(struct sk_buff *skb)
{
	struct rtable *rt = (struct rtable*)skb->dst;
	struct in_device *in_dev = in_dev_get(rt->u.dst.dev);

	if (!in_dev)
		return;

	if (!IN_DEV_TX_REDIRECTS(in_dev))
		goto out;

	/* No redirected packets during ip_rt_redirect_silence;
	 * reset the algorithm.
	 */
	if (jiffies - rt->u.dst.rate_last > ip_rt_redirect_silence)
		rt->u.dst.rate_tokens = 0;

	/* Too many ignored redirects; do not send anything
	 * set u.dst.rate_last to the last seen redirected packet.
	 */
	if (rt->u.dst.rate_tokens >= ip_rt_redirect_number) {
		rt->u.dst.rate_last = jiffies;
		goto out;
	}

	/* Check for load limit; set rate_last to the latest sent
	 * redirect.
	 */
	if (jiffies - rt->u.dst.rate_last > (ip_rt_redirect_load<<rt->u.dst.rate_tokens)) {
		icmp_send(skb, ICMP_REDIRECT, ICMP_REDIR_HOST, rt->rt_gateway);
		rt->u.dst.rate_last = jiffies;
		++rt->u.dst.rate_tokens;
#ifdef CONFIG_IP_ROUTE_VERBOSE
		if (IN_DEV_LOG_MARTIANS(in_dev) &&
		    rt->u.dst.rate_tokens == ip_rt_redirect_number && net_ratelimit())
			printk(KERN_WARNING "host %u.%u.%u.%u/if%d ignores redirects for "
				"%u.%u.%u.%u to %u.%u.%u.%u.\n",
				NIPQUAD(rt->rt_src), rt->rt_iif,
				NIPQUAD(rt->rt_dst), NIPQUAD(rt->rt_gateway));
#endif
	}
out:
        in_dev_put(in_dev);
}

static int ip_error(struct sk_buff *skb)
{
	struct rtable *rt = (struct rtable*)skb->dst;
	unsigned long now;
	int code;

	switch (rt->u.dst.error) {
	case EINVAL:
	default:
		kfree_skb(skb);
		return 0;
	case EHOSTUNREACH:
		code = ICMP_HOST_UNREACH;
		break;
	case ENETUNREACH:
		code = ICMP_NET_UNREACH;
		break;
	case EACCES:
		code = ICMP_PKT_FILTERED;
		break;
	}

	now = jiffies;
	if ((rt->u.dst.rate_tokens += (now - rt->u.dst.rate_last)) > ip_rt_error_burst)
		rt->u.dst.rate_tokens = ip_rt_error_burst;
	rt->u.dst.rate_last = now;
	if (rt->u.dst.rate_tokens >= ip_rt_error_cost) {
		rt->u.dst.rate_tokens -= ip_rt_error_cost;
		icmp_send(skb, ICMP_DEST_UNREACH, code, 0);
	}

	kfree_skb(skb);
	return 0;
} 

/*
 *	The last two values are not from the RFC but
 *	are needed for AMPRnet AX.25 paths.
 */

static unsigned short mtu_plateau[] =
{32000, 17914, 8166, 4352, 2002, 1492, 576, 296, 216, 128 };

static __inline__ unsigned short guess_mtu(unsigned short old_mtu)
{
	int i;
	
	for (i = 0; i < sizeof(mtu_plateau)/sizeof(mtu_plateau[0]); i++)
		if (old_mtu > mtu_plateau[i])
			return mtu_plateau[i];
	return 68;
}

unsigned short ip_rt_frag_needed(struct iphdr *iph, unsigned short new_mtu)
{
	int i;
	unsigned short old_mtu = ntohs(iph->tot_len);
	struct rtable *rth;
	u32  skeys[2] = { iph->saddr, 0, };
	u32  daddr = iph->daddr;
	u8   tos = iph->tos & IPTOS_RT_MASK;
	unsigned short est_mtu = 0;

	if (ipv4_config.no_pmtu_disc)
		return 0;

	for (i=0; i<2; i++) {
		unsigned hash = rt_hash_code(daddr, skeys[i], tos);

		read_lock(&rt_hash_table[hash].lock);
		for (rth = rt_hash_table[hash].chain; rth; rth = rth->u.rt_next) {
			if (rth->key.dst == daddr &&
			    rth->key.src == skeys[i] &&
			    rth->rt_dst == daddr &&
			    rth->rt_src == iph->saddr &&
			    rth->key.tos == tos &&
			    rth->key.iif == 0 &&
			    !(rth->u.dst.mxlock&(1<<RTAX_MTU))) {
				unsigned short mtu = new_mtu;

				if (new_mtu < 68 || new_mtu >= old_mtu) {

					/* BSD 4.2 compatibility hack :-( */
					if (mtu == 0 && old_mtu >= rth->u.dst.pmtu &&
					    old_mtu >= 68 + (iph->ihl<<2))
						old_mtu -= iph->ihl<<2;

					mtu = guess_mtu(old_mtu);
				}
				if (mtu <= rth->u.dst.pmtu) {
					if (mtu < rth->u.dst.pmtu) { 
						dst_confirm(&rth->u.dst);
						if (mtu < ip_rt_min_pmtu) {
							mtu = ip_rt_min_pmtu;
							rth->u.dst.mxlock |= (1<<RTAX_MTU);
						}
						rth->u.dst.pmtu = mtu;
						dst_set_expires(&rth->u.dst, ip_rt_mtu_expires);
					}
					est_mtu = mtu;
				}
			}
		}
		read_unlock(&rt_hash_table[hash].lock);
	}
	return est_mtu ? : new_mtu;
}

void ip_rt_update_pmtu(struct dst_entry *dst, unsigned mtu)
{
	if (dst->pmtu > mtu && mtu >= 68 &&
	    !(dst->mxlock&(1<<RTAX_MTU))) {
		if (mtu < ip_rt_min_pmtu) {
			mtu = ip_rt_min_pmtu;
			dst->mxlock |= (1<<RTAX_MTU);
		}
		dst->pmtu = mtu;
		dst_set_expires(dst, ip_rt_mtu_expires);
	}
}

static struct dst_entry * ipv4_dst_check(struct dst_entry * dst, u32 cookie)
{
	dst_release(dst);
	return NULL;
}

static struct dst_entry * ipv4_dst_reroute(struct dst_entry * dst,
					   struct sk_buff *skb)
{
	return NULL;
}

static void ipv4_dst_destroy(struct dst_entry * dst)
{
	struct rtable *rt = (struct rtable *) dst;
	struct inet_peer *peer = rt->peer;

	if (peer) {
		rt->peer = NULL;
		inet_putpeer(peer);
	}
}

static void ipv4_link_failure(struct sk_buff *skb)
{
	struct rtable *rt;

	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_UNREACH, 0);

	rt = (struct rtable *) skb->dst;
	if (rt)
		dst_set_expires(&rt->u.dst, 0);
}

static int ip_rt_bug(struct sk_buff *skb)
{
	printk(KERN_DEBUG "ip_rt_bug: %u.%u.%u.%u -> %u.%u.%u.%u, %s\n",
		NIPQUAD(skb->nh.iph->saddr), NIPQUAD(skb->nh.iph->daddr),
		skb->dev ? skb->dev->name : "?");
	kfree_skb(skb);
	return 0;
}

/*
   We do not cache source address of outgoing interface,
   because it is used only by IP RR, TS and SRR options,
   so that it out of fast path.

   BTW remember: "addr" is allowed to be not aligned
   in IP options!
 */

void ip_rt_get_source(u8 *addr, struct rtable *rt)
{
	u32 src;
	struct fib_result res;

	if (rt->key.iif == 0)
		src = rt->rt_src;
	else if (fib_lookup(&rt->key, &res) == 0) {
#ifdef CONFIG_IP_ROUTE_NAT
		if (res.type == RTN_NAT)
			src = inet_select_addr(rt->u.dst.dev, rt->rt_gateway, RT_SCOPE_UNIVERSE);
		else
#endif
			src = FIB_RES_PREFSRC(res);
		fib_res_put(&res);
	} else
		src = inet_select_addr(rt->u.dst.dev, rt->rt_gateway, RT_SCOPE_UNIVERSE);
	memcpy(addr, &src, 4);
}

#ifdef CONFIG_NET_CLS_ROUTE
static void set_class_tag(struct rtable *rt, u32 tag)
{
	if (!(rt->u.dst.tclassid&0xFFFF))
		rt->u.dst.tclassid |= tag&0xFFFF;
	if (!(rt->u.dst.tclassid&0xFFFF0000))
		rt->u.dst.tclassid |= tag&0xFFFF0000;
}
#endif

static void rt_set_nexthop(struct rtable *rt, struct fib_result *res, u32 itag)
{
	struct fib_info *fi = res->fi;

	if (fi) {
		if (FIB_RES_GW(*res) && FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
			rt->rt_gateway = FIB_RES_GW(*res);
		memcpy(&rt->u.dst.mxlock, fi->fib_metrics, sizeof(fi->fib_metrics));
		if (fi->fib_mtu == 0) {
			rt->u.dst.pmtu = rt->u.dst.dev->mtu;
			if (rt->u.dst.mxlock&(1<<RTAX_MTU) &&
			    rt->rt_gateway != rt->rt_dst &&
			    rt->u.dst.pmtu > 576)
				rt->u.dst.pmtu = 576;
		}
#ifdef CONFIG_NET_CLS_ROUTE
		rt->u.dst.tclassid = FIB_RES_NH(*res).nh_tclassid;
#endif
	} else {
		rt->u.dst.pmtu	= rt->u.dst.dev->mtu;
	}
	if (rt->u.dst.pmtu > IP_MAX_MTU)
		rt->u.dst.pmtu = IP_MAX_MTU;
	if (rt->u.dst.advmss == 0)
		rt->u.dst.advmss = max(rt->u.dst.dev->mtu-40, ip_rt_min_advmss);
	if (rt->u.dst.advmss > 65535-40)
		rt->u.dst.advmss = 65535-40;

#ifdef CONFIG_NET_CLS_ROUTE
#ifdef CONFIG_IP_MULTIPLE_TABLES
	set_class_tag(rt, fib_rules_tclass(res));
#endif
	set_class_tag(rt, itag);
#endif
        rt->rt_type = res->type;
}

static int
ip_route_input_mc(struct sk_buff *skb, u32 daddr, u32 saddr,
		  u8 tos, struct net_device *dev, int our)
{
	unsigned hash;
	struct rtable *rth;
	u32 spec_dst;
	struct in_device *in_dev = in_dev_get(dev);
	u32 itag = 0;

	/* Primary sanity checks. */

	if (in_dev == NULL)
		return -EINVAL;

	if (MULTICAST(saddr) || BADCLASS(saddr) || LOOPBACK(saddr) ||
	    skb->protocol != __constant_htons(ETH_P_IP))
		goto e_inval;