route.c

嵌入式系统设计与实验教材二源码linux内核移植与编译

{
	struct flow_rule *frule;
	struct rt6_info *nrt = NULL;
	struct pol_chain *pol;

	for (pol = rt6_pol_list; pol; pol = pol->next) {
		struct fib6_node *fn;
		struct rt6_info *sprt;

		fn = fib6_lookup(pol->rules, daddr, saddr);

		do {
			for (sprt = fn->leaf; sprt; sprt=sprt->u.next) {
				int res;

				frule = sprt->rt6i_flowr;
#if RT6_DEBUG >= 2
				if (frule == NULL) {
					printk(KERN_DEBUG "NULL flowr\n");
					goto error;
				}
#endif
				res = frule->ops->accept(rt, sprt, args, &nrt);

				switch (res) {
				case FLOWR_SELECT:
					goto found;
				case FLOWR_CLEAR:
					goto next_policy;
				case FLOWR_NODECISION:
					break;
				default:
					goto error;
				};
			}

			fn = fn->parent;

		} while ((fn->fn_flags & RTN_TL_ROOT) == 0);

	next_policy:
	}

error:
	dst_clone(&ip6_null_entry.u.dst);
	return &ip6_null_entry;

found:
	if (nrt == NULL)
		goto error;

	nrt->rt6i_flags |= RTF_CACHE;
	dst_clone(&nrt->u.dst);
	err = rt6_ins(nrt);
	if (err)
		nrt->u.dst.error = err;
	return nrt;
}
#endif

static int fib6_ifdown(struct rt6_info *rt, void *arg)
{
	if (((void*)rt->rt6i_dev == arg || arg == NULL) &&
	    rt != &ip6_null_entry) {
		RT6_TRACE("deleted by ifdown %p\n", rt);
		return -1;
	}
	return 0;
}

void rt6_ifdown(struct net_device *dev)
{
	write_lock_bh(&rt6_lock);
	fib6_clean_tree(&ip6_routing_table, fib6_ifdown, 0, dev);
	write_unlock_bh(&rt6_lock);
}

struct rt6_mtu_change_arg
{
	struct net_device *dev;
	unsigned mtu;
};

static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg)
{
	struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg;

	/* In IPv6 pmtu discovery is not optional,
	   so that RTAX_MTU lock cannot disable it.
	   We still use this lock to block changes
	   caused by addrconf/ndisc.
	*/
	if (rt->rt6i_dev == arg->dev &&
	    rt->u.dst.pmtu > arg->mtu &&
	    !(rt->u.dst.mxlock&(1<<RTAX_MTU)))
		rt->u.dst.pmtu = arg->mtu;
	rt->u.dst.advmss = max_t(unsigned int, arg->mtu - 60, ip6_rt_min_advmss);
	if (rt->u.dst.advmss > 65535-20)
		rt->u.dst.advmss = 65535;
	return 0;
}

void rt6_mtu_change(struct net_device *dev, unsigned mtu)
{
	struct rt6_mtu_change_arg arg;

	arg.dev = dev;
	arg.mtu = mtu;
	read_lock_bh(&rt6_lock);
	fib6_clean_tree(&ip6_routing_table, rt6_mtu_change_route, 0, &arg);
	read_unlock_bh(&rt6_lock);
}

static int inet6_rtm_to_rtmsg(struct rtmsg *r, struct rtattr **rta,
			      struct in6_rtmsg *rtmsg)
{
	memset(rtmsg, 0, sizeof(*rtmsg));

	rtmsg->rtmsg_dst_len = r->rtm_dst_len;
	rtmsg->rtmsg_src_len = r->rtm_src_len;
	rtmsg->rtmsg_flags = RTF_UP;
	if (r->rtm_type == RTN_UNREACHABLE)
		rtmsg->rtmsg_flags |= RTF_REJECT;

	if (rta[RTA_GATEWAY-1]) {
		if (rta[RTA_GATEWAY-1]->rta_len != RTA_LENGTH(16))
			return -EINVAL;
		memcpy(&rtmsg->rtmsg_gateway, RTA_DATA(rta[RTA_GATEWAY-1]), 16);
		rtmsg->rtmsg_flags |= RTF_GATEWAY;
	}
	if (rta[RTA_DST-1]) {
		if (RTA_PAYLOAD(rta[RTA_DST-1]) < ((r->rtm_dst_len+7)>>3))
			return -EINVAL;
		memcpy(&rtmsg->rtmsg_dst, RTA_DATA(rta[RTA_DST-1]), ((r->rtm_dst_len+7)>>3));
	}
	if (rta[RTA_SRC-1]) {
		if (RTA_PAYLOAD(rta[RTA_SRC-1]) < ((r->rtm_src_len+7)>>3))
			return -EINVAL;
		memcpy(&rtmsg->rtmsg_src, RTA_DATA(rta[RTA_SRC-1]), ((r->rtm_src_len+7)>>3));
	}
	if (rta[RTA_OIF-1]) {
		if (rta[RTA_OIF-1]->rta_len != RTA_LENGTH(sizeof(int)))
			return -EINVAL;
		memcpy(&rtmsg->rtmsg_ifindex, RTA_DATA(rta[RTA_OIF-1]), sizeof(int));
	}
	if (rta[RTA_PRIORITY-1]) {
		if (rta[RTA_PRIORITY-1]->rta_len != RTA_LENGTH(4))
			return -EINVAL;
		memcpy(&rtmsg->rtmsg_metric, RTA_DATA(rta[RTA_PRIORITY-1]), 4);
	}
	return 0;
}

int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
{
	struct rtmsg *r = NLMSG_DATA(nlh);
	struct in6_rtmsg rtmsg;

	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
		return -EINVAL;
	return ip6_route_del(&rtmsg);
}

int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
{
	struct rtmsg *r = NLMSG_DATA(nlh);
	struct in6_rtmsg rtmsg;

	if (inet6_rtm_to_rtmsg(r, arg, &rtmsg))
		return -EINVAL;
	return ip6_route_add(&rtmsg);
}

struct rt6_rtnl_dump_arg
{
	struct sk_buff *skb;
	struct netlink_callback *cb;
};

static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt,
			 struct in6_addr *dst,
			 struct in6_addr *src,
			 int iif,
			 int type, u32 pid, u32 seq)
{
	struct rtmsg *rtm;
	struct nlmsghdr  *nlh;
	unsigned char	 *b = skb->tail;
	struct rta_cacheinfo ci;

	nlh = NLMSG_PUT(skb, pid, seq, type, sizeof(*rtm));
	rtm = NLMSG_DATA(nlh);
	rtm->rtm_family = AF_INET6;
	rtm->rtm_dst_len = rt->rt6i_dst.plen;
	rtm->rtm_src_len = rt->rt6i_src.plen;
	rtm->rtm_tos = 0;
	rtm->rtm_table = RT_TABLE_MAIN;
	if (rt->rt6i_flags&RTF_REJECT)
		rtm->rtm_type = RTN_UNREACHABLE;
	else if (rt->rt6i_dev && (rt->rt6i_dev->flags&IFF_LOOPBACK))
		rtm->rtm_type = RTN_LOCAL;
	else
		rtm->rtm_type = RTN_UNICAST;
	rtm->rtm_flags = 0;
	rtm->rtm_scope = RT_SCOPE_UNIVERSE;
	rtm->rtm_protocol = RTPROT_BOOT;
	if (rt->rt6i_flags&RTF_DYNAMIC)
		rtm->rtm_protocol = RTPROT_REDIRECT;
	else if (rt->rt6i_flags&(RTF_ADDRCONF|RTF_ALLONLINK))
		rtm->rtm_protocol = RTPROT_KERNEL;
	else if (rt->rt6i_flags&RTF_DEFAULT)
		rtm->rtm_protocol = RTPROT_RA;

	if (rt->rt6i_flags&RTF_CACHE)
		rtm->rtm_flags |= RTM_F_CLONED;

	if (dst) {
		RTA_PUT(skb, RTA_DST, 16, dst);
	        rtm->rtm_dst_len = 128;
	} else if (rtm->rtm_dst_len)
		RTA_PUT(skb, RTA_DST, 16, &rt->rt6i_dst.addr);
#ifdef CONFIG_IPV6_SUBTREES
	if (src) {
		RTA_PUT(skb, RTA_SRC, 16, src);
	        rtm->rtm_src_len = 128;
	} else if (rtm->rtm_src_len)
		RTA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr);
#endif
	if (iif)
		RTA_PUT(skb, RTA_IIF, 4, &iif);
	else if (dst) {
		struct in6_addr saddr_buf;
		if (ipv6_get_saddr(&rt->u.dst, dst, &saddr_buf) == 0)
			RTA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf);
	}
	if (rtnetlink_put_metrics(skb, &rt->u.dst.mxlock) < 0)
		goto rtattr_failure;
	if (rt->u.dst.neighbour)
		RTA_PUT(skb, RTA_GATEWAY, 16, &rt->u.dst.neighbour->primary_key);
	if (rt->u.dst.dev)
		RTA_PUT(skb, RTA_OIF, sizeof(int), &rt->rt6i_dev->ifindex);
	RTA_PUT(skb, RTA_PRIORITY, 4, &rt->rt6i_metric);
	ci.rta_lastuse = jiffies - rt->u.dst.lastuse;
	if (rt->rt6i_expires)
		ci.rta_expires = rt->rt6i_expires - jiffies;
	else
		ci.rta_expires = 0;
	ci.rta_used = rt->u.dst.__use;
	ci.rta_clntref = atomic_read(&rt->u.dst.__refcnt);
	ci.rta_error = rt->u.dst.error;
	ci.rta_id = 0;
	ci.rta_ts = 0;
	ci.rta_tsage = 0;
	RTA_PUT(skb, RTA_CACHEINFO, sizeof(ci), &ci);
	nlh->nlmsg_len = skb->tail - b;
	return skb->len;

nlmsg_failure:
rtattr_failure:
	skb_trim(skb, b - skb->data);
	return -1;
}

static int rt6_dump_route(struct rt6_info *rt, void *p_arg)
{
	struct rt6_rtnl_dump_arg *arg = (struct rt6_rtnl_dump_arg *) p_arg;

	return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE,
			     NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq);
}

static int fib6_dump_node(struct fib6_walker_t *w)
{
	int res;
	struct rt6_info *rt;

	for (rt = w->leaf; rt; rt = rt->u.next) {
		res = rt6_dump_route(rt, w->args);
		if (res < 0) {
			/* Frame is full, suspend walking */
			w->leaf = rt;
			return 1;
		}
		BUG_TRAP(res!=0);
	}
	w->leaf = NULL;
	return 0;
}

static void fib6_dump_end(struct netlink_callback *cb)
{
	struct fib6_walker_t *w = (void*)cb->args[0];

	if (w) {
		cb->args[0] = 0;
		fib6_walker_unlink(w);
		kfree(w);
	}
	if (cb->args[1]) {
		cb->done = (void*)cb->args[1];
		cb->args[1] = 0;
	}
}

static int fib6_dump_done(struct netlink_callback *cb)
{
	fib6_dump_end(cb);
	return cb->done(cb);
}

int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct rt6_rtnl_dump_arg arg;
	struct fib6_walker_t *w;
	int res;

	arg.skb = skb;
	arg.cb = cb;

	w = (void*)cb->args[0];
	if (w == NULL) {
		/* New dump:
		 * 
		 * 1. hook callback destructor.
		 */
		cb->args[1] = (long)cb->done;
		cb->done = fib6_dump_done;

		/*
		 * 2. allocate and initialize walker.
		 */
		w = kmalloc(sizeof(*w), GFP_ATOMIC);
		if (w == NULL)
			return -ENOMEM;
		RT6_TRACE("dump<%p", w);
		memset(w, 0, sizeof(*w));
		w->root = &ip6_routing_table;
		w->func = fib6_dump_node;
		w->args = &arg;
		cb->args[0] = (long)w;
		read_lock_bh(&rt6_lock);
		res = fib6_walk(w);
		read_unlock_bh(&rt6_lock);
	} else {
		w->args = &arg;
		read_lock_bh(&rt6_lock);
		res = fib6_walk_continue(w);
		read_unlock_bh(&rt6_lock);
	}
#if RT6_DEBUG >= 3
	if (res <= 0 && skb->len == 0)
		RT6_TRACE("%p>dump end\n", w);
#endif
	res = res < 0 ? res : skb->len;
	/* res < 0 is an error. (really, impossible)
	   res == 0 means that dump is complete, but skb still can contain data.
	   res > 0 dump is not complete, but frame is full.
	 */
	/* Destroy walker, if dump of this table is complete. */
	if (res <= 0)
		fib6_dump_end(cb);
	return res;
}

int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg)
{
	struct rtattr **rta = arg;
	int iif = 0;
	int err;
	struct sk_buff *skb;
	struct flowi fl;
	struct rt6_info *rt;

	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
	if (skb == NULL)
		return -ENOBUFS;

	/* Reserve room for dummy headers, this skb can pass
	   through good chunk of routing engine.
	 */
	skb->mac.raw = skb->data;
	skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr));

	fl.proto = 0;
	fl.nl_u.ip6_u.daddr = NULL;
	fl.nl_u.ip6_u.saddr = NULL;
	fl.uli_u.icmpt.type = 0;
	fl.uli_u.icmpt.code = 0;
	if (rta[RTA_SRC-1])
		fl.nl_u.ip6_u.saddr = (struct in6_addr*)RTA_DATA(rta[RTA_SRC-1]);
	if (rta[RTA_DST-1])
		fl.nl_u.ip6_u.daddr = (struct in6_addr*)RTA_DATA(rta[RTA_DST-1]);

	if (rta[RTA_IIF-1])
		memcpy(&iif, RTA_DATA(rta[RTA_IIF-1]), sizeof(int));

	if (iif) {
		struct net_device *dev;
		dev = __dev_get_by_index(iif);
		if (!dev)
			return -ENODEV;
	}

	fl.oif = 0;
	if (rta[RTA_OIF-1])
		memcpy(&fl.oif, RTA_DATA(rta[RTA_OIF-1]), sizeof(int));

	rt = (struct rt6_info*)ip6_route_output(NULL, &fl);

	skb->dst = &rt->u.dst;

	NETLINK_CB(skb).dst_pid = NETLINK_CB(in_skb).pid;
	err = rt6_fill_node(skb, rt, 
			    fl.nl_u.ip6_u.daddr,
			    fl.nl_u.ip6_u.saddr,
			    iif,
			    RTM_NEWROUTE, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq);
	if (err < 0)
		return -EMSGSIZE;

	err = netlink_unicast(rtnl, skb, NETLINK_CB(in_skb).pid, MSG_DONTWAIT);
	if (err < 0)
		return err;
	return 0;
}

void inet6_rt_notify(int event, struct rt6_info *rt)
{
	struct sk_buff *skb;
	int size = NLMSG_SPACE(sizeof(struct rtmsg)+256);

	skb = alloc_skb(size, gfp_any());
	if (!skb) {
		netlink_set_err(rtnl, 0, RTMGRP_IPV6_ROUTE, ENOBUFS);
		return;
	}
	if (rt6_fill_node(skb, rt, NULL, NULL, 0, event, 0, 0) < 0) {
		kfree_skb(skb);
		netlink_set_err(rtnl, 0, RTMGRP_IPV6_ROUTE, EINVAL);
		return;
	}
	NETLINK_CB(skb).dst_groups = RTMGRP_IPV6_ROUTE;
	netlink_broadcast(rtnl, skb, 0, RTMGRP_IPV6_ROUTE, gfp_any());
}

/*
 *	/proc
 */

#ifdef CONFIG_PROC_FS

#define RT6_INFO_LEN (32 + 4 + 32 + 4 + 32 + 40 + 5 + 1)

struct rt6_proc_arg
{
	char *buffer;
	int offset;
	int length;
	int skip;
	int len;
};

static int rt6_info_route(struct rt6_info *rt, void *p_arg)
{
	struct rt6_proc_arg *arg = (struct rt6_proc_arg *) p_arg;
	int i;

	if (arg->skip < arg->offset / RT6_INFO_LEN) {
		arg->skip++;
		return 0;
	}

	if (arg->len >= arg->length)
		return 0;

	for (i=0; i<16; i++) {
		sprintf(arg->buffer + arg->len, "%02x",
			rt->rt6i_dst.addr.s6_addr[i]);
		arg->len += 2;
	}
	arg->len += sprintf(arg->buffer + arg->len, " %02x ",
			    rt->rt6i_dst.plen);

#ifdef CONFIG_IPV6_SUBTREES
	for (i=0; i<16; i++) {
		sprintf(arg->buffer + arg->len, "%02x",
			rt->rt6i_src.addr.s6_addr[i]);
		arg->len += 2;
	}
	arg->len += sprintf(arg->buffer + arg->len, " %02x ",
			    rt->rt6i_src.plen);
#else
	sprintf(arg->buffer + arg->len,
		"00000000000000000000000000000000 00 ");
	arg->len += 36;
#endif

	if (rt->rt6i_nexthop) {
		for (i=0; i<16; i++) {
			sprintf(arg->buffer + arg->len, "%02x",
				rt->rt6i_nexthop->primary_key[i]);
			arg->len += 2;
		}
	} else {
		sprintf(arg->buffer + arg->len,
			"00000000000000000000000000000000");
		arg->len += 32;
	}
	arg->len += sprintf(arg->buffer + arg->len,
			    " %08x %08x %08x %08x %8s\n",
			    rt->rt6i_metric, atomic_read(&rt->u.dst.__refcnt),
			    rt->u.dst.__use, rt->rt6i_flags, 
			    rt->rt6i_dev ? rt->rt6i_dev->name : "");
	return 0;
}

static int rt6_proc_info(char *buffer, char **start, off_t offset, int length)
{
	struct rt6_proc_arg arg;
	arg.buffer = buffer;
	arg.offset = offset;
	arg.length = length;
	arg.skip = 0;
	arg.len = 0;

	read_lock_bh(&rt6_lock);
	fib6_clean_tree(&ip6_routing_table, rt6_info_route, 0, &arg);
	read_unlock_bh(&rt6_lock);

	*start = buffer;
	if (offset)
		*start += offset % RT6_INFO_LEN;

	arg.len -= offset % RT6_INFO_LEN;

	if (arg.len > length)
		arg.len = length;
	if (arg.len < 0)
		arg.len = 0;

	return arg.len;
}

extern struct rt6_statistics rt6_stats;

static int rt6_proc_stats(char *buffer, char **start, off_t offset, int length)
{
	int len;

	len = sprintf(buffer, "%04x %04x %04x %04x %04x %04x\n",
		      rt6_stats.fib_nodes, rt6_stats.fib_route_nodes,
		      rt6_stats.fib_rt_alloc, rt6_stats.fib_rt_entries,
		      rt6_stats.fib_rt_cache,
		      atomic_read(&ip6_dst_ops.entries));

	len -= offset;

	if (len > length)
		len = length;
	if(len < 0)
		len = 0;

	*start = buffer + offset;

	return len;
}
#endif	/* CONFIG_PROC_FS */

#ifdef CONFIG_SYSCTL

static int flush_delay;

static
int ipv6_sysctl_rtcache_flush(ctl_table *ctl, int write, struct file * filp,
			      void *buffer, size_t *lenp)
{
	if (write) {
		proc_dointvec(ctl, write, filp, buffer, lenp);
		if (flush_delay < 0)
			flush_delay = 0;
		fib6_run_gc((unsigned long)flush_delay);
		return 0;
	} else
		return -EINVAL;
}

ctl_table ipv6_route_table[] = {
        {NET_IPV6_ROUTE_FLUSH, "flush",
         &flush_delay, sizeof(int), 0644, NULL,
         &ipv6_sysctl_rtcache_flush},
	{NET_IPV6_ROUTE_GC_THRESH, "gc_thresh",
         &ip6_dst_ops.gc_thresh, sizeof(int), 0644, NULL,
         &proc_dointvec},
	{NET_IPV6_ROUTE_MAX_SIZE, "max_size",
         &ip6_rt_max_size, sizeof(int), 0644, NULL,
         &proc_dointvec},
	{NET_IPV6_ROUTE_GC_MIN_INTERVAL, "gc_min_interval",
         &ip6_rt_gc_min_interval, sizeof(int), 0644, NULL,
         &proc_dointvec_jiffies, &sysctl_jiffies},
	{NET_IPV6_ROUTE_GC_TIMEOUT, "gc_timeout",
         &ip6_rt_gc_timeout, sizeof(int), 0644, NULL,
         &proc_dointvec_jiffies, &sysctl_jiffies},
	{NET_IPV6_ROUTE_GC_INTERVAL, "gc_interval",
         &ip6_rt_gc_interval, sizeof(int), 0644, NULL,
         &proc_dointvec_jiffies, &sysctl_jiffies},
	{NET_IPV6_ROUTE_GC_ELASTICITY, "gc_elasticity",
         &ip6_rt_gc_elasticity, sizeof(int), 0644, NULL,
         &proc_dointvec_jiffies, &sysctl_jiffies},
	{NET_IPV6_ROUTE_MTU_EXPIRES, "mtu_expires",
         &ip6_rt_mtu_expires, sizeof(int), 0644, NULL,
         &proc_dointvec_jiffies, &sysctl_jiffies},
	{NET_IPV6_ROUTE_MIN_ADVMSS, "min_adv_mss",
         &ip6_rt_min_advmss, sizeof(int), 0644, NULL,
         &proc_dointvec_jiffies, &sysctl_jiffies},
	 {0}
};

#endif


void __init ip6_route_init(void)
{
	ip6_dst_ops.kmem_cachep = kmem_cache_create("ip6_dst_cache",
						     sizeof(struct rt6_info),
						     0, SLAB_HWCACHE_ALIGN,
						     NULL, NULL);
	fib6_init();
#ifdef 	CONFIG_PROC_FS
	proc_net_create("ipv6_route", 0, rt6_proc_info);
	proc_net_create("rt6_stats", 0, rt6_proc_stats);
#endif
}

#ifdef MODULE
void ip6_route_cleanup(void)
{
#ifdef CONFIG_PROC_FS
	proc_net_remove("ipv6_route");
	proc_net_remove("rt6_stats");
#endif

	rt6_ifdown(NULL);
	fib6_gc_cleanup();
}
#endif	/* MODULE */