patch-2.1.68 linux/net/ipv4/fib_frontend.c

Next file: linux/net/ipv4/fib_hash.c
Previous file: linux/net/ipv4/fib.c
Back to the patch index
Back to the overall index

diff -u --recursive --new-file v2.1.67/linux/net/ipv4/fib_frontend.c linux/net/ipv4/fib_frontend.c
@@ -0,0 +1,572 @@
+/*
+ * INET		An implementation of the TCP/IP protocol suite for the LINUX
+ *		operating system.  INET is implemented using the  BSD Socket
+ *		interface as the means of communication with the user level.
+ *
+ *		IPv4 Forwarding Information Base: FIB frontend.
+ *
+ * Version:	$Id: fib_frontend.c,v 1.4 1997/11/09 20:05:23 kuznet Exp $
+ *
+ * Authors:	Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/config.h>
+#include <asm/uaccess.h>
+#include <asm/system.h>
+#include <asm/bitops.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/string.h>
+#include <linux/socket.h>
+#include <linux/sockios.h>
+#include <linux/errno.h>
+#include <linux/in.h>
+#include <linux/inet.h>
+#include <linux/netdevice.h>
+#include <linux/if_arp.h>
+#include <linux/proc_fs.h>
+#include <linux/skbuff.h>
+#include <linux/netlink.h>
+#include <linux/init.h>
+
+#include <net/ip.h>
+#include <net/protocol.h>
+#include <net/route.h>
+#include <net/tcp.h>
+#include <net/sock.h>
+#include <net/icmp.h>
+#include <net/arp.h>
+#include <net/ip_fib.h>
+
+#define FFprint(a...) printk(KERN_DEBUG a)
+
+#ifndef CONFIG_IP_MULTIPLE_TABLES
+
+#define RT_TABLE_MIN RT_TABLE_MAIN
+
+struct fib_table *local_table;
+struct fib_table *main_table;
+
+#else
+
+#define RT_TABLE_MIN 1
+
+struct fib_table *fib_tables[RT_TABLE_MAX+1];
+
+struct fib_table *__fib_new_table(int id)
+{
+	struct fib_table *tb;
+
+	tb = fib_hash_init(id);
+	if (!tb)
+		return NULL;
+	fib_tables[id] = tb;
+	return tb;
+}
+
+
+#endif /* CONFIG_IP_MULTIPLE_TABLES */
+
+
+void fib_flush(void)
+{
+	int flushed = 0;
+#ifdef CONFIG_IP_MULTIPLE_TABLES
+	struct fib_table *tb;
+	int id;
+
+	for (id = RT_TABLE_MAX; id>0; id--) {
+		if ((tb = fib_get_table(id))==NULL)
+			continue;
+		flushed += tb->tb_flush(tb);
+	}
+#else /* CONFIG_IP_MULTIPLE_TABLES */
+	flushed += main_table->tb_flush(main_table);
+	flushed += local_table->tb_flush(local_table);
+#endif /* CONFIG_IP_MULTIPLE_TABLES */
+
+	if (flushed)
+		rt_cache_flush(RT_FLUSH_DELAY);
+}
+
+
+#ifdef CONFIG_PROC_FS
+
+/* 
+ *	Called from the PROCfs module. This outputs /proc/net/route.
+ *
+ *	It always works in backward compatibility mode.
+ *	The format of the file is not supposed to be changed.
+ */
+ 
+static int
+fib_get_procinfo(char *buffer, char **start, off_t offset, int length, int dummy)
+{
+	int first = offset/128;
+	char *ptr = buffer;
+	int count = (length+127)/128;
+	int len;
+
+	*start = buffer + offset%128;
+	
+	if (--first < 0) {
+		sprintf(buffer, "%-127s\n", "Iface\tDestination\tGateway \tFlags\tRefCnt\tUse\tMetric\tMask\t\tMTU\tWindow\tIRTT");
+		--count;
+		ptr += 128;
+		first = 0;
+  	}
+
+	/* rtnl_shlock(); -- it is pointless at the moment --ANK */
+	if (main_table && count > 0) {
+		int n = main_table->tb_get_info(main_table, ptr, first, count);
+		count -= n;
+		ptr += n*128;
+	}
+	/* rtnl_shunlock(); */
+	len = ptr - *start;
+	if (len >= length)
+		return length;
+	if (len >= 0)
+		return len;
+	return 0;
+}
+
+#endif /* CONFIG_PROC_FS */
+
+/*
+ *	Find the first device with a given source address.
+ */
+
+struct device * ip_dev_find(u32 addr)
+{
+	struct rt_key key;
+	struct fib_result res;
+
+	memset(&key, 0, sizeof(key));
+	key.dst = addr;
+	key.scope = RT_SCOPE_UNIVERSE;
+
+	if (!local_table || local_table->tb_lookup(local_table, &key, &res)
+	    || res.type != RTN_LOCAL)
+		return NULL;
+
+	return FIB_RES_DEV(res);
+}
+
+unsigned inet_addr_type(u32 addr)
+{
+	struct rt_key		key;
+	struct fib_result	res;
+
+	if (ZERONET(addr) || BADCLASS(addr))
+		return RTN_BROADCAST;
+	if (MULTICAST(addr))
+		return RTN_MULTICAST;
+
+	memset(&key, 0, sizeof(key));
+	key.dst = addr;
+
+	if (local_table) {
+		if (local_table->tb_lookup(local_table, &key, &res) == 0)
+			return res.type;
+		return RTN_UNICAST;
+	}
+	return RTN_BROADCAST;
+}
+
+/* Given (packet source, input interface) and optional (dst, oif, tos):
+   - (main) check, that source is valid i.e. not broadcast or our local
+     address.
+   - figure out what "logical" interface this packet arrived
+     and calculate "specific destination" address.
+   - check, that packet arrived from expected physical interface.
+ */
+
+int fib_validate_source(u32 src, u32 dst, u8 tos, int oif,
+			struct device *dev, u32 *spec_dst)
+{
+	struct in_device *in_dev = dev->ip_ptr;
+	struct rt_key key;
+	struct fib_result res;
+
+	key.dst = src;
+	key.src = dst;
+	key.tos = tos;
+	key.oif = 0;
+	key.iif = oif;
+	key.scope = RT_SCOPE_UNIVERSE;
+
+	if (in_dev == NULL)
+		return -EINVAL;
+	if (fib_lookup(&key, &res))
+		goto last_resort;
+	if (res.type != RTN_UNICAST)
+		return -EINVAL;
+	*spec_dst = FIB_RES_PREFSRC(res);
+#ifdef CONFIG_IP_ROUTE_MULTIPATH
+	if (FIB_RES_DEV(res) == dev || res.fi->fib_nhs > 1)
+#else
+	if (FIB_RES_DEV(res) == dev)
+#endif
+		return FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
+
+	if (in_dev->ifa_list == NULL)
+		goto last_resort;
+	if (IN_DEV_RPFILTER(in_dev))
+		return -EINVAL;
+	key.oif = dev->ifindex;
+	if (fib_lookup(&key, &res) == 0 && res.type == RTN_UNICAST) {
+		*spec_dst = FIB_RES_PREFSRC(res);
+		return FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
+	}
+	return 0;
+
+last_resort:
+	if (IN_DEV_RPFILTER(in_dev))
+		return -EINVAL;
+	*spec_dst = inet_select_addr(dev, 0, RT_SCOPE_UNIVERSE);
+	return 0;
+}
+
+#ifndef CONFIG_IP_NOSIOCRT
+
+/*
+ *	Handle IP routing ioctl calls. These are used to manipulate the routing tables
+ */
+ 
+int ip_rt_ioctl(unsigned int cmd, void *arg)
+{
+	int err;
+	struct kern_rta rta;
+	struct rtentry  r;
+	struct {
+		struct nlmsghdr nlh;
+		struct rtmsg	rtm;
+	} req;
+
+	switch (cmd) {
+	case SIOCADDRT:		/* Add a route */
+	case SIOCDELRT:		/* Delete a route */
+		if (!suser())
+			return -EPERM;
+		if (copy_from_user(&r, arg, sizeof(struct rtentry)))
+			return -EFAULT;
+		rtnl_lock();
+		err = fib_convert_rtentry(cmd, &req.nlh, &req.rtm, &rta, arg);
+		if (err == 0) {
+			if (cmd == SIOCDELRT) {
+				struct fib_table *tb = fib_get_table(req.rtm.rtm_table);
+				err = -ESRCH;
+				if (tb)
+					err = tb->tb_delete(tb, &req.rtm, &rta, &req.nlh, NULL);
+			} else {
+				struct fib_table *tb = fib_new_table(req.rtm.rtm_table);
+				err = -ENOBUFS;
+				if (tb)
+					err = tb->tb_insert(tb, &req.rtm, &rta, &req.nlh, NULL);
+			}
+		}
+		rtnl_unlock();
+		return err;
+	}
+	return -EINVAL;
+}
+
+#else
+
+int ip_rt_ioctl(unsigned int cmd, void *arg)
+{
+	return -EINVAL;
+}
+
+#endif
+
+#ifdef CONFIG_RTNETLINK
+
+int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
+{
+	struct fib_table * tb;
+	struct kern_rta *rta = arg;
+	struct rtmsg *r = NLMSG_DATA(nlh);
+
+	tb = fib_get_table(r->rtm_table);
+	if (tb)
+		return tb->tb_delete(tb, r, rta, nlh, &NETLINK_CB(skb));
+	return -ESRCH;
+}
+
+int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
+{
+	struct fib_table * tb;
+	struct kern_rta *rta = arg;
+	struct rtmsg *r = NLMSG_DATA(nlh);
+
+	tb = fib_new_table(r->rtm_table);
+	if (tb)
+		return tb->tb_insert(tb, r, rta, nlh, &NETLINK_CB(skb));
+	return -ENOBUFS;
+}
+
+int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
+{
+	int t;
+	int s_t;
+	struct fib_table *tb;
+
+	s_t = cb->args[0];
+	if (s_t == 0)
+		s_t = cb->args[0] = RT_TABLE_MIN;
+
+	for (t=s_t; t<=RT_TABLE_MAX; t++) {
+		if (t < s_t) continue;
+		if (t > s_t)
+			memset(&cb->args[1], 0, sizeof(cb->args)-sizeof(int));
+		if ((tb = fib_get_table(t))==NULL)
+			continue;
+		if (tb->tb_dump(tb, skb, cb) < 0) 
+			break;
+	}
+
+	cb->args[0] = t;
+
+	return skb->len;
+}
+
+#endif
+
+/* Prepare and feed intra-kernel routing request.
+   Really, it should be netlink message, but :-( netlink
+   can be not configured, so that we feed it directly
+   to fib engine. It is legal, because all events occur
+   only when netlink is already locked.
+ */
+
+static void fib_magic(int cmd, int type, u32 dst, int dst_len, struct in_ifaddr *ifa)
+{
+	struct fib_table * tb;
+	struct {
+		struct nlmsghdr	nlh;
+		struct rtmsg	rtm;
+	} req;
+	struct kern_rta rta;
+
+	memset(&req.rtm, 0, sizeof(req.rtm));
+	memset(&rta, 0, sizeof(rta));
+
+	if (type == RTN_UNICAST)
+		tb = fib_new_table(RT_TABLE_MAIN);
+	else
+		tb = fib_new_table(RT_TABLE_LOCAL);
+
+	if (tb == NULL)
+		return;
+
+	req.nlh.nlmsg_len = sizeof(req);
+	req.nlh.nlmsg_type = cmd;
+	req.nlh.nlmsg_flags = NLM_F_REQUEST|NLM_F_CREATE;
+	req.nlh.nlmsg_pid = 0;
+	req.nlh.nlmsg_seq = 0;
+
+	req.rtm.rtm_dst_len = dst_len;
+	req.rtm.rtm_table = tb->tb_id;
+	req.rtm.rtm_protocol = RTPROT_KERNEL;
+	req.rtm.rtm_scope = (type != RTN_LOCAL ? RT_SCOPE_LINK : RT_SCOPE_HOST);
+	req.rtm.rtm_type = type;
+
+	rta.rta_dst = &dst;
+	rta.rta_prefsrc = &ifa->ifa_local;
+	rta.rta_oif = &ifa->ifa_dev->dev->ifindex;
+
+	if (cmd == RTM_NEWROUTE)
+		tb->tb_insert(tb, &req.rtm, &rta, &req.nlh, NULL);
+	else
+		tb->tb_delete(tb, &req.rtm, &rta, &req.nlh, NULL);
+}
+
+static void fib_add_ifaddr(struct in_ifaddr *ifa)
+{
+	struct in_device *in_dev = ifa->ifa_dev;
+	struct device *dev = in_dev->dev;
+	struct in_ifaddr *prim = ifa;
+	u32 mask = ifa->ifa_mask;
+	u32 addr = ifa->ifa_local;
+	u32 prefix = ifa->ifa_address&mask;
+
+	if (ifa->ifa_flags&IFA_F_SECONDARY)
+		prim = inet_ifa_byprefix(in_dev, prefix, mask);
+
+	fib_magic(RTM_NEWROUTE, RTN_LOCAL, addr, 32, prim);
+
+	if (!(dev->flags&IFF_UP))
+		return;
+
+	/* Add broadcast address, if it is explicitly assigned. */
+	if (ifa->ifa_broadcast && ifa->ifa_broadcast != 0xFFFFFFFF)
+		fib_magic(RTM_NEWROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim);
+
+	if (!ZERONET(prefix) && !(ifa->ifa_flags&IFA_F_SECONDARY)) {
+		fib_magic(RTM_NEWROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL :
+			  RTN_UNICAST, prefix, ifa->ifa_prefixlen, prim);
+
+		/* Add network specific broadcasts, when it takes a sense */
+		if (ifa->ifa_prefixlen < 31) {
+			fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix, 32, prim);
+			fib_magic(RTM_NEWROUTE, RTN_BROADCAST, prefix|~mask, 32, prim);
+		}
+	}
+}
+
+static void fib_del_ifaddr(struct in_ifaddr *ifa)
+{
+	struct in_device *in_dev = ifa->ifa_dev;
+	struct device *dev = in_dev->dev;
+	struct in_ifaddr *ifa1;
+	struct in_ifaddr *prim = ifa;
+	u32 brd = ifa->ifa_address|~ifa->ifa_mask;
+	u32 any = ifa->ifa_address&ifa->ifa_mask;
+#define LOCAL_OK	1
+#define BRD_OK		2
+#define BRD0_OK		4
+#define BRD1_OK		8
+	unsigned ok = 0;
+
+	if (!(ifa->ifa_flags&IFA_F_SECONDARY))
+		fib_magic(RTM_DELROUTE, dev->flags&IFF_LOOPBACK ? RTN_LOCAL :
+			  RTN_UNICAST, any, ifa->ifa_prefixlen, prim);
+	else 
+		prim = inet_ifa_byprefix(in_dev, any, ifa->ifa_mask);
+
+	/* Deletion is more complicated than add.
+	   We should take care of not to delete too much :-)
+
+	   Scan address list to be sure that addresses are really gone.
+	 */
+
+	for (ifa1 = in_dev->ifa_list; ifa1; ifa1 = ifa1->ifa_next) {
+		if (ifa->ifa_local == ifa1->ifa_local)
+			ok |= LOCAL_OK;
+		if (ifa->ifa_broadcast == ifa1->ifa_broadcast)
+			ok |= BRD_OK;
+		if (brd == ifa1->ifa_broadcast)
+			ok |= BRD1_OK;
+		if (any == ifa1->ifa_broadcast)
+			ok |= BRD0_OK;
+	}
+
+	if (!(ok&BRD_OK))
+		fib_magic(RTM_DELROUTE, RTN_BROADCAST, ifa->ifa_broadcast, 32, prim);
+	if (!(ok&BRD1_OK))
+		fib_magic(RTM_DELROUTE, RTN_BROADCAST, brd, 32, prim);
+	if (!(ok&BRD0_OK))
+		fib_magic(RTM_DELROUTE, RTN_BROADCAST, any, 32, prim);
+	if (!(ok&LOCAL_OK)) {
+		fib_magic(RTM_DELROUTE, RTN_LOCAL, ifa->ifa_local, 32, prim);
+
+		/* Check, that this local address finally disappeared. */
+		if (inet_addr_type(ifa->ifa_local) != RTN_LOCAL) {
+			/* And the last, but not the least thing.
+			   We must flush stray FIB entries.
+
+			   First of all, we scan fib_info list searching
+			   for stray nexthop entries, then ignite fib_flush.
+			*/
+			if (fib_sync_down(ifa->ifa_local, NULL))
+				fib_flush();
+		}
+	}
+#undef LOCAL_OK
+#undef BRD_OK
+#undef BRD0_OK
+#undef BRD1_OK
+}
+
+static int fib_inetaddr_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+	struct in_ifaddr *ifa = (struct in_ifaddr*)ptr;
+
+	switch (event) {
+	case NETDEV_UP:
+		fib_add_ifaddr(ifa);
+		rt_cache_flush(2*HZ);
+		break;
+	case NETDEV_DOWN:
+		fib_del_ifaddr(ifa);
+		rt_cache_flush(1*HZ);
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+static int fib_netdev_event(struct notifier_block *this, unsigned long event, void *ptr)
+{
+	struct device *dev = ptr;
+	struct in_device *in_dev = dev->ip_ptr;
+
+	if (!in_dev)
+		return NOTIFY_DONE;
+
+	switch (event) {
+	case NETDEV_UP:
+		for_ifa(in_dev) {
+			fib_add_ifaddr(ifa);
+		} endfor_ifa(in_dev);
+#ifdef CONFIG_IP_ROUTE_MULTIPATH
+		fib_sync_up(dev);
+#endif
+		rt_cache_flush(2*HZ);
+		break;
+	case NETDEV_DOWN:
+		if (fib_sync_down(0, dev))
+			fib_flush();
+		rt_cache_flush(0);
+		break;
+	case NETDEV_UNREGISTER:
+		if (in_dev->ifa_list)
+			printk("About to crash!\n");
+		rt_cache_flush(0);
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+struct notifier_block fib_inetaddr_notifier = {
+	fib_inetaddr_event,
+	NULL,
+	0
+};
+
+struct notifier_block fib_netdev_notifier = {
+	fib_netdev_event,
+	NULL,
+	0
+};
+
+__initfunc(void ip_fib_init(void))
+{
+#ifdef CONFIG_PROC_FS
+	proc_net_register(&(struct proc_dir_entry) {
+		PROC_NET_ROUTE, 5, "route",
+		S_IFREG | S_IRUGO, 1, 0, 0,
+		0, &proc_net_inode_operations,
+		fib_get_procinfo
+	});
+#endif		/* CONFIG_PROC_FS */
+
+#ifndef CONFIG_IP_MULTIPLE_TABLES
+	local_table = fib_hash_init(RT_TABLE_LOCAL);
+	main_table = fib_hash_init(RT_TABLE_MAIN);
+#else
+	fib_rules_init();
+#endif
+
+	register_netdevice_notifier(&fib_netdev_notifier);
+	register_inetaddr_notifier(&fib_inetaddr_notifier);
+}
+

FUNET's LINUX-ADM group, linux-adm@nic.funet.fi
TCL-scripts by Sam Shen, slshen@lbl.gov