[net-next-2.6.git] / net / core / rtnetlink.c

/*
 * 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.
 *
 *		Routing netlink socket interface: protocol independent part.
 *
 * 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.
 *
 *	Fixes:
 *	Vitaly E. Lavrov		RTA_OK arithmetics was wrong.
 */

#include <linux/config.h>
#include <linux/errno.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/timer.h>
#include <linux/string.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/fcntl.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/capability.h>
#include <linux/skbuff.h>
#include <linux/init.h>
#include <linux/security.h>

#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/string.h>

#include <linux/inet.h>
#include <linux/netdevice.h>
#include <net/ip.h>
#include <net/protocol.h>
#include <net/arp.h>
#include <net/route.h>
#include <net/udp.h>
#include <net/sock.h>
#include <net/pkt_sched.h>

DECLARE_MUTEX(rtnl_sem);

void rtnl_lock(void)
{
	rtnl_shlock();
}

int rtnl_lock_interruptible(void)
{
	return down_interruptible(&rtnl_sem);
}
 
void rtnl_unlock(void)
{
	rtnl_shunlock();

	netdev_run_todo();
}

int rtattr_parse(struct rtattr *tb[], int maxattr, struct rtattr *rta, int len)
{
	memset(tb, 0, sizeof(struct rtattr*)*maxattr);

	while (RTA_OK(rta, len)) {
		unsigned flavor = rta->rta_type;
		if (flavor && flavor <= maxattr)
			tb[flavor-1] = rta;
		rta = RTA_NEXT(rta, len);
	}
	return 0;
}

struct sock *rtnl;

struct rtnetlink_link * rtnetlink_links[NPROTO];

static const int rtm_min[RTM_NR_FAMILIES] =
{
	[RTM_FAM(RTM_NEWLINK)]      = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
	[RTM_FAM(RTM_NEWADDR)]      = NLMSG_LENGTH(sizeof(struct ifaddrmsg)),
	[RTM_FAM(RTM_NEWROUTE)]     = NLMSG_LENGTH(sizeof(struct rtmsg)),
	[RTM_FAM(RTM_NEWNEIGH)]     = NLMSG_LENGTH(sizeof(struct ndmsg)),
	[RTM_FAM(RTM_NEWRULE)]      = NLMSG_LENGTH(sizeof(struct rtmsg)),
	[RTM_FAM(RTM_NEWQDISC)]     = NLMSG_LENGTH(sizeof(struct tcmsg)),
	[RTM_FAM(RTM_NEWTCLASS)]    = NLMSG_LENGTH(sizeof(struct tcmsg)),
	[RTM_FAM(RTM_NEWTFILTER)]   = NLMSG_LENGTH(sizeof(struct tcmsg)),
	[RTM_FAM(RTM_NEWACTION)]    = NLMSG_LENGTH(sizeof(struct tcamsg)),
	[RTM_FAM(RTM_NEWPREFIX)]    = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	[RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	[RTM_FAM(RTM_GETANYCAST)]   = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	[RTM_FAM(RTM_NEWNEIGHTBL)]  = NLMSG_LENGTH(sizeof(struct ndtmsg)),
};

static const int rta_max[RTM_NR_FAMILIES] =
{
	[RTM_FAM(RTM_NEWLINK)]      = IFLA_MAX,
	[RTM_FAM(RTM_NEWADDR)]      = IFA_MAX,
	[RTM_FAM(RTM_NEWROUTE)]     = RTA_MAX,
	[RTM_FAM(RTM_NEWNEIGH)]     = NDA_MAX,
	[RTM_FAM(RTM_NEWRULE)]      = RTA_MAX,
	[RTM_FAM(RTM_NEWQDISC)]     = TCA_MAX,
	[RTM_FAM(RTM_NEWTCLASS)]    = TCA_MAX,
	[RTM_FAM(RTM_NEWTFILTER)]   = TCA_MAX,
	[RTM_FAM(RTM_NEWACTION)]    = TCAA_MAX,
	[RTM_FAM(RTM_NEWNEIGHTBL)]  = NDTA_MAX,
};

void __rta_fill(struct sk_buff *skb, int attrtype, int attrlen, const void *data)
{
	struct rtattr *rta;
	int size = RTA_LENGTH(attrlen);

	rta = (struct rtattr*)skb_put(skb, RTA_ALIGN(size));
	rta->rta_type = attrtype;
	rta->rta_len = size;
	memcpy(RTA_DATA(rta), data, attrlen);
	memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size);
}

size_t rtattr_strlcpy(char *dest, const struct rtattr *rta, size_t size)
{
	size_t ret = RTA_PAYLOAD(rta);
	char *src = RTA_DATA(rta);

	if (ret > 0 && src[ret - 1] == '\0')
		ret--;
	if (size > 0) {
		size_t len = (ret >= size) ? size - 1 : ret;
		memset(dest, 0, size);
		memcpy(dest, src, len);
	}
	return ret;
}

int rtnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo)
{
	int err = 0;

	NETLINK_CB(skb).dst_group = group;
	if (echo)
		atomic_inc(&skb->users);
	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	if (echo)
		err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	return err;
}

int rtnetlink_put_metrics(struct sk_buff *skb, u32 *metrics)
{
	struct rtattr *mx = (struct rtattr*)skb->tail;
	int i;

	RTA_PUT(skb, RTA_METRICS, 0, NULL);
	for (i=0; i<RTAX_MAX; i++) {
		if (metrics[i])
			RTA_PUT(skb, i+1, sizeof(u32), metrics+i);
	}
	mx->rta_len = skb->tail - (u8*)mx;
	if (mx->rta_len == RTA_LENGTH(0))
		skb_trim(skb, (u8*)mx - skb->data);
	return 0;

rtattr_failure:
	skb_trim(skb, (u8*)mx - skb->data);
	return -1;
}


static int rtnetlink_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
				 int type, u32 pid, u32 seq, u32 change, 
				 unsigned int flags)
{
	struct ifinfomsg *r;
	struct nlmsghdr  *nlh;
	unsigned char	 *b = skb->tail;

	nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*r), flags);
	r = NLMSG_DATA(nlh);
	r->ifi_family = AF_UNSPEC;
	r->__ifi_pad = 0;
	r->ifi_type = dev->type;
	r->ifi_index = dev->ifindex;
	r->ifi_flags = dev_get_flags(dev);
	r->ifi_change = change;

	RTA_PUT(skb, IFLA_IFNAME, strlen(dev->name)+1, dev->name);

	if (1) {
		u32 txqlen = dev->tx_queue_len;
		RTA_PUT(skb, IFLA_TXQLEN, sizeof(txqlen), &txqlen);
	}

	if (1) {
		u32 weight = dev->weight;
		RTA_PUT(skb, IFLA_WEIGHT, sizeof(weight), &weight);
	}

	if (1) {
		struct rtnl_link_ifmap map = {
			.mem_start   = dev->mem_start,
			.mem_end     = dev->mem_end,
			.base_addr   = dev->base_addr,
			.irq         = dev->irq,
			.dma         = dev->dma,
			.port        = dev->if_port,
		};
		RTA_PUT(skb, IFLA_MAP, sizeof(map), &map);
	}

	if (dev->addr_len) {
		RTA_PUT(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr);
		RTA_PUT(skb, IFLA_BROADCAST, dev->addr_len, dev->broadcast);
	}

	if (1) {
		u32 mtu = dev->mtu;
		RTA_PUT(skb, IFLA_MTU, sizeof(mtu), &mtu);
	}

	if (dev->ifindex != dev->iflink) {
		u32 iflink = dev->iflink;
		RTA_PUT(skb, IFLA_LINK, sizeof(iflink), &iflink);
	}

	if (dev->qdisc_sleeping)
		RTA_PUT(skb, IFLA_QDISC,
			strlen(dev->qdisc_sleeping->ops->id) + 1,
			dev->qdisc_sleeping->ops->id);
	
	if (dev->master) {
		u32 master = dev->master->ifindex;
		RTA_PUT(skb, IFLA_MASTER, sizeof(master), &master);
	}

	if (dev->get_stats) {
		unsigned long *stats = (unsigned long*)dev->get_stats(dev);
		if (stats) {
			struct rtattr  *a;
			__u32	       *s;
			int		i;
			int		n = sizeof(struct rtnl_link_stats)/4;

			a = __RTA_PUT(skb, IFLA_STATS, n*4);
			s = RTA_DATA(a);
			for (i=0; i<n; i++)
				s[i] = stats[i];
		}
	}
	nlh->nlmsg_len = skb->tail - b;
	return skb->len;

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

static int rtnetlink_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
{
	int idx;
	int s_idx = cb->args[0];
	struct net_device *dev;

	read_lock(&dev_base_lock);
	for (dev=dev_base, idx=0; dev; dev = dev->next, idx++) {
		if (idx < s_idx)
			continue;
		if (rtnetlink_fill_ifinfo(skb, dev, RTM_NEWLINK,
					  NETLINK_CB(cb->skb).pid,
					  cb->nlh->nlmsg_seq, 0,
					  NLM_F_MULTI) <= 0)
			break;
	}
	read_unlock(&dev_base_lock);
	cb->args[0] = idx;

	return skb->len;
}

static int do_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
	struct ifinfomsg  *ifm = NLMSG_DATA(nlh);
	struct rtattr    **ida = arg;
	struct net_device *dev;
	int err, send_addr_notify = 0;

	if (ifm->ifi_index >= 0)
		dev = dev_get_by_index(ifm->ifi_index);
	else if (ida[IFLA_IFNAME - 1]) {
		char ifname[IFNAMSIZ];

		if (rtattr_strlcpy(ifname, ida[IFLA_IFNAME - 1],
		                   IFNAMSIZ) >= IFNAMSIZ)
			return -EINVAL;
		dev = dev_get_by_name(ifname);
	} else
		return -EINVAL;

	if (!dev)
		return -ENODEV;

	err = -EINVAL;

	if (ifm->ifi_flags)
		dev_change_flags(dev, ifm->ifi_flags);

	if (ida[IFLA_MAP - 1]) {
		struct rtnl_link_ifmap *u_map;
		struct ifmap k_map;

		if (!dev->set_config) {
			err = -EOPNOTSUPP;
			goto out;
		}

		if (!netif_device_present(dev)) {
			err = -ENODEV;
			goto out;
		}
		
		if (ida[IFLA_MAP - 1]->rta_len != RTA_LENGTH(sizeof(*u_map)))
			goto out;

		u_map = RTA_DATA(ida[IFLA_MAP - 1]);

		k_map.mem_start = (unsigned long) u_map->mem_start;
		k_map.mem_end = (unsigned long) u_map->mem_end;
		k_map.base_addr = (unsigned short) u_map->base_addr;
		k_map.irq = (unsigned char) u_map->irq;
		k_map.dma = (unsigned char) u_map->dma;
		k_map.port = (unsigned char) u_map->port;

		err = dev->set_config(dev, &k_map);

		if (err)
			goto out;
	}

	if (ida[IFLA_ADDRESS - 1]) {
		if (!dev->set_mac_address) {
			err = -EOPNOTSUPP;
			goto out;
		}
		if (!netif_device_present(dev)) {
			err = -ENODEV;
			goto out;
		}
		if (ida[IFLA_ADDRESS - 1]->rta_len != RTA_LENGTH(dev->addr_len))
			goto out;

		err = dev->set_mac_address(dev, RTA_DATA(ida[IFLA_ADDRESS - 1]));
		if (err)
			goto out;
		send_addr_notify = 1;
	}

	if (ida[IFLA_BROADCAST - 1]) {
		if (ida[IFLA_BROADCAST - 1]->rta_len != RTA_LENGTH(dev->addr_len))
			goto out;
		memcpy(dev->broadcast, RTA_DATA(ida[IFLA_BROADCAST - 1]),
		       dev->addr_len);
		send_addr_notify = 1;
	}

	if (ida[IFLA_MTU - 1]) {
		if (ida[IFLA_MTU - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
			goto out;
		err = dev_set_mtu(dev, *((u32 *) RTA_DATA(ida[IFLA_MTU - 1])));

		if (err)
			goto out;

	}

	if (ida[IFLA_TXQLEN - 1]) {
		if (ida[IFLA_TXQLEN - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
			goto out;

		dev->tx_queue_len = *((u32 *) RTA_DATA(ida[IFLA_TXQLEN - 1]));
	}

	if (ida[IFLA_WEIGHT - 1]) {
		if (ida[IFLA_WEIGHT - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
			goto out;

		dev->weight = *((u32 *) RTA_DATA(ida[IFLA_WEIGHT - 1]));
	}

	if (ifm->ifi_index >= 0 && ida[IFLA_IFNAME - 1]) {
		char ifname[IFNAMSIZ];

		if (rtattr_strlcpy(ifname, ida[IFLA_IFNAME - 1],
		                   IFNAMSIZ) >= IFNAMSIZ)
			goto out;
		err = dev_change_name(dev, ifname);
		if (err)
			goto out;
	}

	err = 0;

out:
	if (send_addr_notify)
		call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);

	dev_put(dev);
	return err;
}

static int rtnetlink_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
{
	int idx;
	int s_idx = cb->family;

	if (s_idx == 0)
		s_idx = 1;
	for (idx=1; idx<NPROTO; idx++) {
		int type = cb->nlh->nlmsg_type-RTM_BASE;
		if (idx < s_idx || idx == PF_PACKET)
			continue;
		if (rtnetlink_links[idx] == NULL ||
		    rtnetlink_links[idx][type].dumpit == NULL)
			continue;
		if (idx > s_idx)
			memset(&cb->args[0], 0, sizeof(cb->args));
		if (rtnetlink_links[idx][type].dumpit(skb, cb))
			break;
	}
	cb->family = idx;

	return skb->len;
}

void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
{
	struct sk_buff *skb;
	int size = NLMSG_SPACE(sizeof(struct ifinfomsg) +
			       sizeof(struct rtnl_link_ifmap) +
			       sizeof(struct rtnl_link_stats) + 128);

	skb = alloc_skb(size, GFP_KERNEL);
	if (!skb)
		return;

	if (rtnetlink_fill_ifinfo(skb, dev, type, current->pid, 0, change, 0) < 0) {
		kfree_skb(skb);
		return;
	}
	NETLINK_CB(skb).dst_group = RTNLGRP_LINK;
	netlink_broadcast(rtnl, skb, 0, RTNLGRP_LINK, GFP_KERNEL);
}

/* Protected by RTNL sempahore.  */
static struct rtattr **rta_buf;
static int rtattr_max;

/* Process one rtnetlink message. */

static __inline__ int
rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, int *errp)
{
	struct rtnetlink_link *link;
	struct rtnetlink_link *link_tab;
	int sz_idx, kind;
	int min_len;
	int family;
	int type;
	int err;

	/* Only requests are handled by kernel now */
	if (!(nlh->nlmsg_flags&NLM_F_REQUEST))
		return 0;

	type = nlh->nlmsg_type;

	/* A control message: ignore them */
	if (type < RTM_BASE)
		return 0;

	/* Unknown message: reply with EINVAL */
	if (type > RTM_MAX)
		goto err_inval;

	type -= RTM_BASE;

	/* All the messages must have at least 1 byte length */
	if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(struct rtgenmsg)))
		return 0;

	family = ((struct rtgenmsg*)NLMSG_DATA(nlh))->rtgen_family;
	if (family >= NPROTO) {
		*errp = -EAFNOSUPPORT;
		return -1;
	}

	link_tab = rtnetlink_links[family];
	if (link_tab == NULL)
		link_tab = rtnetlink_links[PF_UNSPEC];
	link = &link_tab[type];

	sz_idx = type>>2;
	kind = type&3;

	if (kind != 2 && security_netlink_recv(skb)) {
		*errp = -EPERM;
		return -1;
	}

	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
		u32 rlen;

		if (link->dumpit == NULL)
			link = &(rtnetlink_links[PF_UNSPEC][type]);

		if (link->dumpit == NULL)
			goto err_inval;

		if ((*errp = netlink_dump_start(rtnl, skb, nlh,
						link->dumpit, NULL)) != 0) {
			return -1;
		}
		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
		if (rlen > skb->len)
			rlen = skb->len;
		skb_pull(skb, rlen);
		return -1;
	}

	memset(rta_buf, 0, (rtattr_max * sizeof(struct rtattr *)));

	min_len = rtm_min[sz_idx];
	if (nlh->nlmsg_len < min_len)
		goto err_inval;

	if (nlh->nlmsg_len > min_len) {
		int attrlen = nlh->nlmsg_len - NLMSG_ALIGN(min_len);
		struct rtattr *attr = (void*)nlh + NLMSG_ALIGN(min_len);

		while (RTA_OK(attr, attrlen)) {
			unsigned flavor = attr->rta_type;
			if (flavor) {
				if (flavor > rta_max[sz_idx])
					goto err_inval;
				rta_buf[flavor-1] = attr;
			}
			attr = RTA_NEXT(attr, attrlen);
		}
	}

	if (link->doit == NULL)
		link = &(rtnetlink_links[PF_UNSPEC][type]);
	if (link->doit == NULL)
		goto err_inval;
	err = link->doit(skb, nlh, (void *)&rta_buf[0]);

	*errp = err;
	return err;

err_inval:
	*errp = -EINVAL;
	return -1;
}

/* 
 * Process one packet of messages.
 * Malformed skbs with wrong lengths of messages are discarded silently.
 */

static inline int rtnetlink_rcv_skb(struct sk_buff *skb)
{
	int err;
	struct nlmsghdr * nlh;

	while (skb->len >= NLMSG_SPACE(0)) {
		u32 rlen;

		nlh = (struct nlmsghdr *)skb->data;
		if (nlh->nlmsg_len < sizeof(*nlh) || skb->len < nlh->nlmsg_len)
			return 0;
		rlen = NLMSG_ALIGN(nlh->nlmsg_len);
		if (rlen > skb->len)
			rlen = skb->len;
		if (rtnetlink_rcv_msg(skb, nlh, &err)) {
			/* Not error, but we must interrupt processing here:
			 *   Note, that in this case we do not pull message
			 *   from skb, it will be processed later.
			 */
			if (err == 0)
				return -1;
			netlink_ack(skb, nlh, err);
		} else if (nlh->nlmsg_flags&NLM_F_ACK)
			netlink_ack(skb, nlh, 0);
		skb_pull(skb, rlen);
	}

	return 0;
}

/*
 *  rtnetlink input queue processing routine:
 *	- process as much as there was in the queue upon entry.
 *	- feed skbs to rtnetlink_rcv_skb, until it refuse a message,
 *	  that will occur, when a dump started.
 */

static void rtnetlink_rcv(struct sock *sk, int len)
{
	unsigned int qlen = skb_queue_len(&sk->sk_receive_queue);

	do {
		struct sk_buff *skb;

		rtnl_lock();

		if (qlen > skb_queue_len(&sk->sk_receive_queue))
			qlen = skb_queue_len(&sk->sk_receive_queue);

		for (; qlen; qlen--) {
			skb = skb_dequeue(&sk->sk_receive_queue);
			if (rtnetlink_rcv_skb(skb)) {
				if (skb->len)
					skb_queue_head(&sk->sk_receive_queue,
						       skb);
				else {
					kfree_skb(skb);
					qlen--;
				}
				break;
			}
			kfree_skb(skb);
		}

		up(&rtnl_sem);

		netdev_run_todo();
	} while (qlen);
}

static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] =
{
	[RTM_GETLINK     - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo },
	[RTM_SETLINK     - RTM_BASE] = { .doit   = do_setlink		 },
	[RTM_GETADDR     - RTM_BASE] = { .dumpit = rtnetlink_dump_all	 },
	[RTM_GETROUTE    - RTM_BASE] = { .dumpit = rtnetlink_dump_all	 },
	[RTM_NEWNEIGH    - RTM_BASE] = { .doit   = neigh_add		 },
	[RTM_DELNEIGH    - RTM_BASE] = { .doit   = neigh_delete		 },
	[RTM_GETNEIGH    - RTM_BASE] = { .dumpit = neigh_dump_info	 },
	[RTM_GETRULE     - RTM_BASE] = { .dumpit = rtnetlink_dump_all	 },
	[RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info	 },
	[RTM_SETNEIGHTBL - RTM_BASE] = { .doit   = neightbl_set		 },
};

static int rtnetlink_event(struct notifier_block *this, unsigned long event, void *ptr)
{
	struct net_device *dev = ptr;
	switch (event) {
	case NETDEV_UNREGISTER:
		rtmsg_ifinfo(RTM_DELLINK, dev, ~0U);
		break;
	case NETDEV_REGISTER:
		rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U);
		break;
	case NETDEV_UP:
	case NETDEV_DOWN:
		rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP|IFF_RUNNING);
		break;
	case NETDEV_CHANGE:
	case NETDEV_GOING_DOWN:
		break;
	default:
		rtmsg_ifinfo(RTM_NEWLINK, dev, 0);
		break;
	}
	return NOTIFY_DONE;
}

static struct notifier_block rtnetlink_dev_notifier = {
	.notifier_call	= rtnetlink_event,
};

void __init rtnetlink_init(void)
{
	int i;

	rtattr_max = 0;
	for (i = 0; i < ARRAY_SIZE(rta_max); i++)
		if (rta_max[i] > rtattr_max)
			rtattr_max = rta_max[i];
	rta_buf = kmalloc(rtattr_max * sizeof(struct rtattr *), GFP_KERNEL);
	if (!rta_buf)
		panic("rtnetlink_init: cannot allocate rta_buf\n");

	rtnl = netlink_kernel_create(NETLINK_ROUTE, RTNLGRP_MAX, rtnetlink_rcv,
	                             THIS_MODULE);
	if (rtnl == NULL)
		panic("rtnetlink_init: cannot initialize rtnetlink\n");
	netlink_set_nonroot(NETLINK_ROUTE, NL_NONROOT_RECV);
	register_netdevice_notifier(&rtnetlink_dev_notifier);
	rtnetlink_links[PF_UNSPEC] = link_rtnetlink_table;
	rtnetlink_links[PF_PACKET] = link_rtnetlink_table;
}

EXPORT_SYMBOL(__rta_fill);
EXPORT_SYMBOL(rtattr_strlcpy);
EXPORT_SYMBOL(rtattr_parse);
EXPORT_SYMBOL(rtnetlink_links);
EXPORT_SYMBOL(rtnetlink_put_metrics);
EXPORT_SYMBOL(rtnl);
EXPORT_SYMBOL(rtnl_lock);
EXPORT_SYMBOL(rtnl_lock_interruptible);
EXPORT_SYMBOL(rtnl_sem);
EXPORT_SYMBOL(rtnl_unlock);
Commit	Line	Data
1da177e4 LT	1	/*
	2	* INET An implementation of the TCP/IP protocol suite for the LINUX
	3	* operating system. INET is implemented using the BSD Socket
	4	* interface as the means of communication with the user level.
	5	*
	6	* Routing netlink socket interface: protocol independent part.
	7	*
	8	* Authors: Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
	9	*
	10	* This program is free software; you can redistribute it and/or
	11	* modify it under the terms of the GNU General Public License
	12	* as published by the Free Software Foundation; either version
	13	* 2 of the License, or (at your option) any later version.
	14	*
	15	* Fixes:
	16	* Vitaly E. Lavrov RTA_OK arithmetics was wrong.
	17	*/
	18
	19	#include <linux/config.h>
	20	#include <linux/errno.h>
	21	#include <linux/module.h>
	22	#include <linux/types.h>
	23	#include <linux/socket.h>
	24	#include <linux/kernel.h>
1da177e4 LT	25	#include <linux/sched.h>
	26	#include <linux/timer.h>
	27	#include <linux/string.h>
	28	#include <linux/sockios.h>
	29	#include <linux/net.h>
	30	#include <linux/fcntl.h>
	31	#include <linux/mm.h>
	32	#include <linux/slab.h>
	33	#include <linux/interrupt.h>
	34	#include <linux/capability.h>
	35	#include <linux/skbuff.h>
	36	#include <linux/init.h>
	37	#include <linux/security.h>
	38
	39	#include <asm/uaccess.h>
	40	#include <asm/system.h>
	41	#include <asm/string.h>
	42
	43	#include <linux/inet.h>
	44	#include <linux/netdevice.h>
	45	#include <net/ip.h>
	46	#include <net/protocol.h>
	47	#include <net/arp.h>
	48	#include <net/route.h>
	49	#include <net/udp.h>
	50	#include <net/sock.h>
	51	#include <net/pkt_sched.h>
	52
	53	DECLARE_MUTEX(rtnl_sem);
	54
	55	void rtnl_lock(void)
	56	{
	57	rtnl_shlock();
	58	}
	59
	60	int rtnl_lock_interruptible(void)
	61	{
	62	return down_interruptible(&rtnl_sem);
	63	}
	64
	65	void rtnl_unlock(void)
	66	{
	67	rtnl_shunlock();
	68
	69	netdev_run_todo();
	70	}
	71
	72	int rtattr_parse(struct rtattr tb[], int maxattr, struct rtattr rta, int len)
	73	{
	74	memset(tb, 0, sizeof(struct rtattr)maxattr);
	75
	76	while (RTA_OK(rta, len)) {
	77	unsigned flavor = rta->rta_type;
	78	if (flavor && flavor <= maxattr)
	79	tb[flavor-1] = rta;
	80	rta = RTA_NEXT(rta, len);
	81	}
	82	return 0;
	83	}
	84
	85	struct sock *rtnl;
	86
	87	struct rtnetlink_link * rtnetlink_links[NPROTO];
	88
db46edc6	89	static const int rtm_min[RTM_NR_FAMILIES] =
1da177e4	90	{
f90a0a74 TG	91	[RTM_FAM(RTM_NEWLINK)] = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
	92	[RTM_FAM(RTM_NEWADDR)] = NLMSG_LENGTH(sizeof(struct ifaddrmsg)),
	93	[RTM_FAM(RTM_NEWROUTE)] = NLMSG_LENGTH(sizeof(struct rtmsg)),
	94	[RTM_FAM(RTM_NEWNEIGH)] = NLMSG_LENGTH(sizeof(struct ndmsg)),
	95	[RTM_FAM(RTM_NEWRULE)] = NLMSG_LENGTH(sizeof(struct rtmsg)),
	96	[RTM_FAM(RTM_NEWQDISC)] = NLMSG_LENGTH(sizeof(struct tcmsg)),
	97	[RTM_FAM(RTM_NEWTCLASS)] = NLMSG_LENGTH(sizeof(struct tcmsg)),
	98	[RTM_FAM(RTM_NEWTFILTER)] = NLMSG_LENGTH(sizeof(struct tcmsg)),
	99	[RTM_FAM(RTM_NEWACTION)] = NLMSG_LENGTH(sizeof(struct tcamsg)),
	100	[RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	101	[RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	102	[RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
c7fb64db	103	[RTM_FAM(RTM_NEWNEIGHTBL)] = NLMSG_LENGTH(sizeof(struct ndtmsg)),
1da177e4 LT	104	};
1da177e4 LT	105
db46edc6	106	static const int rta_max[RTM_NR_FAMILIES] =
1da177e4	107	{
f90a0a74 TG	108	[RTM_FAM(RTM_NEWLINK)] = IFLA_MAX,
	109	[RTM_FAM(RTM_NEWADDR)] = IFA_MAX,
	110	[RTM_FAM(RTM_NEWROUTE)] = RTA_MAX,
	111	[RTM_FAM(RTM_NEWNEIGH)] = NDA_MAX,
	112	[RTM_FAM(RTM_NEWRULE)] = RTA_MAX,
	113	[RTM_FAM(RTM_NEWQDISC)] = TCA_MAX,
	114	[RTM_FAM(RTM_NEWTCLASS)] = TCA_MAX,
	115	[RTM_FAM(RTM_NEWTFILTER)] = TCA_MAX,
	116	[RTM_FAM(RTM_NEWACTION)] = TCAA_MAX,
c7fb64db	117	[RTM_FAM(RTM_NEWNEIGHTBL)] = NDTA_MAX,
1da177e4 LT	118	};
	119
	120	void __rta_fill(struct sk_buff skb, int attrtype, int attrlen, const void data)
	121	{
	122	struct rtattr *rta;
	123	int size = RTA_LENGTH(attrlen);
	124
	125	rta = (struct rtattr*)skb_put(skb, RTA_ALIGN(size));
	126	rta->rta_type = attrtype;
	127	rta->rta_len = size;
	128	memcpy(RTA_DATA(rta), data, attrlen);
b3563c4f	129	memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size);
1da177e4 LT	130	}
	131
	132	size_t rtattr_strlcpy(char dest, const struct rtattr rta, size_t size)
	133	{
	134	size_t ret = RTA_PAYLOAD(rta);
	135	char *src = RTA_DATA(rta);
	136
	137	if (ret > 0 && src[ret - 1] == '\0')
	138	ret--;
	139	if (size > 0) {
	140	size_t len = (ret >= size) ? size - 1 : ret;
	141	memset(dest, 0, size);
	142	memcpy(dest, src, len);
	143	}
	144	return ret;
	145	}
	146
	147	int rtnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo)
	148	{
	149	int err = 0;
	150
ac6d439d	151	NETLINK_CB(skb).dst_group = group;
1da177e4 LT	152	if (echo)
	153	atomic_inc(&skb->users);
	154	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	155	if (echo)
	156	err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	157	return err;
	158	}
	159
	160	int rtnetlink_put_metrics(struct sk_buff skb, u32 metrics)
	161	{
	162	struct rtattr mx = (struct rtattr)skb->tail;
	163	int i;
	164
	165	RTA_PUT(skb, RTA_METRICS, 0, NULL);
	166	for (i=0; i<RTAX_MAX; i++) {
	167	if (metrics[i])
	168	RTA_PUT(skb, i+1, sizeof(u32), metrics+i);
	169	}
	170	mx->rta_len = skb->tail - (u8*)mx;
	171	if (mx->rta_len == RTA_LENGTH(0))
	172	skb_trim(skb, (u8*)mx - skb->data);
	173	return 0;
	174
	175	rtattr_failure:
	176	skb_trim(skb, (u8*)mx - skb->data);
	177	return -1;
	178	}
	179
	180
	181	static int rtnetlink_fill_ifinfo(struct sk_buff skb, struct net_device dev,
b6544c0b JHS	182	int type, u32 pid, u32 seq, u32 change,
b6544c0b JHS	183	unsigned int flags)
1da177e4 LT	184	{
	185	struct ifinfomsg *r;
	186	struct nlmsghdr *nlh;
	187	unsigned char *b = skb->tail;
	188
b6544c0b	189	nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*r), flags);
1da177e4 LT	190	r = NLMSG_DATA(nlh);
1da177e4 LT	191	r->ifi_family = AF_UNSPEC;
9ef1d4c7	192	r->__ifi_pad = 0;
1da177e4 LT	193	r->ifi_type = dev->type;
	194	r->ifi_index = dev->ifindex;
	195	r->ifi_flags = dev_get_flags(dev);
	196	r->ifi_change = change;
	197
	198	RTA_PUT(skb, IFLA_IFNAME, strlen(dev->name)+1, dev->name);
	199
	200	if (1) {
	201	u32 txqlen = dev->tx_queue_len;
	202	RTA_PUT(skb, IFLA_TXQLEN, sizeof(txqlen), &txqlen);
	203	}
	204
	205	if (1) {
	206	u32 weight = dev->weight;
	207	RTA_PUT(skb, IFLA_WEIGHT, sizeof(weight), &weight);
	208	}
	209
	210	if (1) {
	211	struct rtnl_link_ifmap map = {
	212	.mem_start = dev->mem_start,
	213	.mem_end = dev->mem_end,
	214	.base_addr = dev->base_addr,
	215	.irq = dev->irq,
	216	.dma = dev->dma,
	217	.port = dev->if_port,
	218	};
	219	RTA_PUT(skb, IFLA_MAP, sizeof(map), &map);
	220	}
	221
	222	if (dev->addr_len) {
	223	RTA_PUT(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr);
	224	RTA_PUT(skb, IFLA_BROADCAST, dev->addr_len, dev->broadcast);
	225	}
	226
	227	if (1) {
	228	u32 mtu = dev->mtu;
	229	RTA_PUT(skb, IFLA_MTU, sizeof(mtu), &mtu);
	230	}
	231
	232	if (dev->ifindex != dev->iflink) {
	233	u32 iflink = dev->iflink;
	234	RTA_PUT(skb, IFLA_LINK, sizeof(iflink), &iflink);
	235	}
	236
	237	if (dev->qdisc_sleeping)
	238	RTA_PUT(skb, IFLA_QDISC,
	239	strlen(dev->qdisc_sleeping->ops->id) + 1,
	240	dev->qdisc_sleeping->ops->id);
	241
	242	if (dev->master) {
	243	u32 master = dev->master->ifindex;
	244	RTA_PUT(skb, IFLA_MASTER, sizeof(master), &master);
	245	}
	246
	247	if (dev->get_stats) {
	248	unsigned long stats = (unsigned long)dev->get_stats(dev);
	249	if (stats) {
	250	struct rtattr *a;
	251	__u32 *s;
	252	int i;
	253	int n = sizeof(struct rtnl_link_stats)/4;
	254
	255	a = __RTA_PUT(skb, IFLA_STATS, n*4);
	256	s = RTA_DATA(a);
257	for (i=0; i<n; i++)
258	s[i] = stats[i];
259	}
260	}
261	nlh->nlmsg_len = skb->tail - b;
262	return skb->len;
263
264	nlmsg_failure:
265	rtattr_failure:
266	skb_trim(skb, b - skb->data);
267	return -1;
268	}
269
270	static int rtnetlink_dump_ifinfo(struct sk_buff skb, struct netlink_callback cb)
271	{
272	int idx;
273	int s_idx = cb->args[0];
274	struct net_device *dev;
275
276	read_lock(&dev_base_lock);
277	for (dev=dev_base, idx=0; dev; dev = dev->next, idx++) {
278	if (idx < s_idx)
279	continue;
b6544c0b JHS	280	if (rtnetlink_fill_ifinfo(skb, dev, RTM_NEWLINK,
	281	NETLINK_CB(cb->skb).pid,
	282	cb->nlh->nlmsg_seq, 0,
	283	NLM_F_MULTI) <= 0)
1da177e4 LT	284	break;
	285	}
	286	read_unlock(&dev_base_lock);
	287	cb->args[0] = idx;
	288
	289	return skb->len;
	290	}
	291
	292	static int do_setlink(struct sk_buff skb, struct nlmsghdr nlh, void *arg)
	293	{
	294	struct ifinfomsg *ifm = NLMSG_DATA(nlh);
	295	struct rtattr **ida = arg;
	296	struct net_device *dev;
	297	int err, send_addr_notify = 0;
	298
	299	if (ifm->ifi_index >= 0)
	300	dev = dev_get_by_index(ifm->ifi_index);
	301	else if (ida[IFLA_IFNAME - 1]) {
	302	char ifname[IFNAMSIZ];
	303
	304	if (rtattr_strlcpy(ifname, ida[IFLA_IFNAME - 1],
	305	IFNAMSIZ) >= IFNAMSIZ)
	306	return -EINVAL;
	307	dev = dev_get_by_name(ifname);
	308	} else
	309	return -EINVAL;
	310
	311	if (!dev)
	312	return -ENODEV;
	313
	314	err = -EINVAL;
	315
	316	if (ifm->ifi_flags)
	317	dev_change_flags(dev, ifm->ifi_flags);
	318
	319	if (ida[IFLA_MAP - 1]) {
	320	struct rtnl_link_ifmap *u_map;
	321	struct ifmap k_map;
	322
	323	if (!dev->set_config) {
	324	err = -EOPNOTSUPP;
	325	goto out;
	326	}
	327
	328	if (!netif_device_present(dev)) {
	329	err = -ENODEV;
	330	goto out;
	331	}
	332
	333	if (ida[IFLA_MAP - 1]->rta_len != RTA_LENGTH(sizeof(*u_map)))
	334	goto out;
	335
	336	u_map = RTA_DATA(ida[IFLA_MAP - 1]);
	337
	338	k_map.mem_start = (unsigned long) u_map->mem_start;
	339	k_map.mem_end = (unsigned long) u_map->mem_end;
	340	k_map.base_addr = (unsigned short) u_map->base_addr;
	341	k_map.irq = (unsigned char) u_map->irq;
	342	k_map.dma = (unsigned char) u_map->dma;
	343	k_map.port = (unsigned char) u_map->port;
	344
	345	err = dev->set_config(dev, &k_map);
	346
	347	if (err)
348	goto out;
349	}
350
351	if (ida[IFLA_ADDRESS - 1]) {
352	if (!dev->set_mac_address) {
353	err = -EOPNOTSUPP;
354	goto out;
355	}
356	if (!netif_device_present(dev)) {
357	err = -ENODEV;
358	goto out;
359	}
360	if (ida[IFLA_ADDRESS - 1]->rta_len != RTA_LENGTH(dev->addr_len))
361	goto out;
362
363	err = dev->set_mac_address(dev, RTA_DATA(ida[IFLA_ADDRESS - 1]));
364	if (err)
365	goto out;
366	send_addr_notify = 1;
367	}
368
369	if (ida[IFLA_BROADCAST - 1]) {
370	if (ida[IFLA_BROADCAST - 1]->rta_len != RTA_LENGTH(dev->addr_len))
371	goto out;
372	memcpy(dev->broadcast, RTA_DATA(ida[IFLA_BROADCAST - 1]),
373	dev->addr_len);
374	send_addr_notify = 1;
375	}
376
377	if (ida[IFLA_MTU - 1]) {
378	if (ida[IFLA_MTU - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
379	goto out;
380	err = dev_set_mtu(dev, ((u32 ) RTA_DATA(ida[IFLA_MTU - 1])));
381
382	if (err)
383	goto out;
384
385	}
386
387	if (ida[IFLA_TXQLEN - 1]) {
388	if (ida[IFLA_TXQLEN - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
389	goto out;
390
391	dev->tx_queue_len = ((u32 ) RTA_DATA(ida[IFLA_TXQLEN - 1]));
392	}
393
394	if (ida[IFLA_WEIGHT - 1]) {
395	if (ida[IFLA_WEIGHT - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
396	goto out;
397
398	dev->weight = ((u32 ) RTA_DATA(ida[IFLA_WEIGHT - 1]));
399	}
400
401	if (ifm->ifi_index >= 0 && ida[IFLA_IFNAME - 1]) {
402	char ifname[IFNAMSIZ];
403
404	if (rtattr_strlcpy(ifname, ida[IFLA_IFNAME - 1],
405	IFNAMSIZ) >= IFNAMSIZ)
406	goto out;
407	err = dev_change_name(dev, ifname);
408	if (err)
409	goto out;
410	}
411
412	err = 0;
413
414	out:
415	if (send_addr_notify)
416	call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
417
418	dev_put(dev);
419	return err;
420	}
421
422	static int rtnetlink_dump_all(struct sk_buff skb, struct netlink_callback cb)
423	{
424	int idx;
425	int s_idx = cb->family;
426
427	if (s_idx == 0)
428	s_idx = 1;
429	for (idx=1; idx<NPROTO; idx++) {
430	int type = cb->nlh->nlmsg_type-RTM_BASE;
431	if (idx < s_idx \|\| idx == PF_PACKET)
432	continue;
433	if (rtnetlink_links[idx] == NULL \|\|
434	rtnetlink_links[idx][type].dumpit == NULL)
435	continue;
436	if (idx > s_idx)
437	memset(&cb->args[0], 0, sizeof(cb->args));
438	if (rtnetlink_links[idx][type].dumpit(skb, cb))
439	break;
440	}
441	cb->family = idx;
442
443	return skb->len;
444	}
445
446	void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
447	{
448	struct sk_buff *skb;
449	int size = NLMSG_SPACE(sizeof(struct ifinfomsg) +
450	sizeof(struct rtnl_link_ifmap) +
451	sizeof(struct rtnl_link_stats) + 128);
452
453	skb = alloc_skb(size, GFP_KERNEL);
454	if (!skb)
455	return;
456
9ed19f33	457	if (rtnetlink_fill_ifinfo(skb, dev, type, current->pid, 0, change, 0) < 0) {
1da177e4 LT	458	kfree_skb(skb);
	459	return;
	460	}
ac6d439d PM	461	NETLINK_CB(skb).dst_group = RTNLGRP_LINK;
ac6d439d PM	462	netlink_broadcast(rtnl, skb, 0, RTNLGRP_LINK, GFP_KERNEL);
1da177e4 LT	463	}
1da177e4 LT	464
1da177e4 LT	465	/* Protected by RTNL sempahore. */
	466	static struct rtattr **rta_buf;
	467	static int rtattr_max;
	468
	469	/* Process one rtnetlink message. */
	470
	471	static __inline__ int
	472	rtnetlink_rcv_msg(struct sk_buff skb, struct nlmsghdr nlh, int *errp)
	473	{
	474	struct rtnetlink_link *link;
	475	struct rtnetlink_link *link_tab;
	476	int sz_idx, kind;
	477	int min_len;
	478	int family;
	479	int type;
	480	int err;
	481
	482	/* Only requests are handled by kernel now */
	483	if (!(nlh->nlmsg_flags&NLM_F_REQUEST))
	484	return 0;
	485
	486	type = nlh->nlmsg_type;
	487
	488	/* A control message: ignore them */
	489	if (type < RTM_BASE)
	490	return 0;
	491
	492	/* Unknown message: reply with EINVAL */
	493	if (type > RTM_MAX)
	494	goto err_inval;
	495
	496	type -= RTM_BASE;
	497
	498	/* All the messages must have at least 1 byte length */
	499	if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(struct rtgenmsg)))
	500	return 0;
	501
	502	family = ((struct rtgenmsg*)NLMSG_DATA(nlh))->rtgen_family;
	503	if (family >= NPROTO) {
	504	*errp = -EAFNOSUPPORT;
	505	return -1;
	506	}
	507
	508	link_tab = rtnetlink_links[family];
	509	if (link_tab == NULL)
	510	link_tab = rtnetlink_links[PF_UNSPEC];
	511	link = &link_tab[type];
	512
	513	sz_idx = type>>2;
	514	kind = type&3;
	515
	516	if (kind != 2 && security_netlink_recv(skb)) {
	517	*errp = -EPERM;
	518	return -1;
	519	}
	520
	521	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
	522	u32 rlen;
	523
	524	if (link->dumpit == NULL)
	525	link = &(rtnetlink_links[PF_UNSPEC][type]);
	526
	527	if (link->dumpit == NULL)
	528	goto err_inval;
529
530	if ((*errp = netlink_dump_start(rtnl, skb, nlh,
a8f74b22	531	link->dumpit, NULL)) != 0) {
1da177e4 LT	532	return -1;
	533	}
	534	rlen = NLMSG_ALIGN(nlh->nlmsg_len);
	535	if (rlen > skb->len)
	536	rlen = skb->len;
	537	skb_pull(skb, rlen);
	538	return -1;
	539	}
	540
	541	memset(rta_buf, 0, (rtattr_max * sizeof(struct rtattr *)));
	542
	543	min_len = rtm_min[sz_idx];
	544	if (nlh->nlmsg_len < min_len)
	545	goto err_inval;
	546
	547	if (nlh->nlmsg_len > min_len) {
	548	int attrlen = nlh->nlmsg_len - NLMSG_ALIGN(min_len);
	549	struct rtattr attr = (void)nlh + NLMSG_ALIGN(min_len);
	550
	551	while (RTA_OK(attr, attrlen)) {
	552	unsigned flavor = attr->rta_type;
	553	if (flavor) {
	554	if (flavor > rta_max[sz_idx])
	555	goto err_inval;
	556	rta_buf[flavor-1] = attr;
	557	}
	558	attr = RTA_NEXT(attr, attrlen);
	559	}
	560	}
	561
	562	if (link->doit == NULL)
	563	link = &(rtnetlink_links[PF_UNSPEC][type]);
	564	if (link->doit == NULL)
	565	goto err_inval;
	566	err = link->doit(skb, nlh, (void *)&rta_buf[0]);
	567
	568	*errp = err;
	569	return err;
	570
	571	err_inval:
	572	*errp = -EINVAL;
	573	return -1;
	574	}
	575
	576	/*
	577	* Process one packet of messages.
	578	* Malformed skbs with wrong lengths of messages are discarded silently.
	579	*/
	580
	581	static inline int rtnetlink_rcv_skb(struct sk_buff *skb)
	582	{
	583	int err;
	584	struct nlmsghdr * nlh;
	585
	586	while (skb->len >= NLMSG_SPACE(0)) {
	587	u32 rlen;
	588
	589	nlh = (struct nlmsghdr *)skb->data;
	590	if (nlh->nlmsg_len < sizeof(*nlh) \|\| skb->len < nlh->nlmsg_len)
	591	return 0;
	592	rlen = NLMSG_ALIGN(nlh->nlmsg_len);
	593	if (rlen > skb->len)
	594	rlen = skb->len;
	595	if (rtnetlink_rcv_msg(skb, nlh, &err)) {
596	/* Not error, but we must interrupt processing here:
597	* Note, that in this case we do not pull message
598	* from skb, it will be processed later.
599	*/
600	if (err == 0)
601	return -1;
602	netlink_ack(skb, nlh, err);
603	} else if (nlh->nlmsg_flags&NLM_F_ACK)
604	netlink_ack(skb, nlh, 0);
605	skb_pull(skb, rlen);
606	}
607
608	return 0;
609	}
610
611	/*
612	* rtnetlink input queue processing routine:
2a0a6ebe	613	* - process as much as there was in the queue upon entry.
1da177e4	614	* - feed skbs to rtnetlink_rcv_skb, until it refuse a message,
2a0a6ebe	615	* that will occur, when a dump started.
1da177e4 LT	616	*/
	617
	618	static void rtnetlink_rcv(struct sock *sk, int len)
	619	{
2a0a6ebe HX	620	unsigned int qlen = skb_queue_len(&sk->sk_receive_queue);
2a0a6ebe HX	621
1da177e4 LT	622	do {
	623	struct sk_buff *skb;
	624
2a0a6ebe HX	625	rtnl_lock();
	626
	627	if (qlen > skb_queue_len(&sk->sk_receive_queue))
	628	qlen = skb_queue_len(&sk->sk_receive_queue);
1da177e4	629
09e14305	630	for (; qlen; qlen--) {
2a0a6ebe	631	skb = skb_dequeue(&sk->sk_receive_queue);
1da177e4	632	if (rtnetlink_rcv_skb(skb)) {
09e14305	633	if (skb->len)
1da177e4 LT	634	skb_queue_head(&sk->sk_receive_queue,
1da177e4 LT	635	skb);
0f4821e7	636	else {
1da177e4	637	kfree_skb(skb);
0f4821e7 DM	638	qlen--;
0f4821e7 DM	639	}
1da177e4 LT	640	break;
	641	}
	642	kfree_skb(skb);
	643	}
	644
	645	up(&rtnl_sem);
	646
	647	netdev_run_todo();
2a0a6ebe	648	} while (qlen);
1da177e4 LT	649	}
1da177e4 LT	650
db46edc6	651	static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] =
1da177e4	652	{
c7fb64db TG	653	[RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo },
	654	[RTM_SETLINK - RTM_BASE] = { .doit = do_setlink },
	655	[RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all },
	656	[RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all },
	657	[RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add },
	658	[RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete },
	659	[RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info },
	660	[RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all },
	661	[RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info },
	662	[RTM_SETNEIGHTBL - RTM_BASE] = { .doit = neightbl_set },
1da177e4 LT	663	};
	664
	665	static int rtnetlink_event(struct notifier_block this, unsigned long event, void ptr)
	666	{
	667	struct net_device *dev = ptr;
	668	switch (event) {
	669	case NETDEV_UNREGISTER:
	670	rtmsg_ifinfo(RTM_DELLINK, dev, ~0U);
	671	break;
	672	case NETDEV_REGISTER:
	673	rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U);
	674	break;
	675	case NETDEV_UP:
	676	case NETDEV_DOWN:
	677	rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP\|IFF_RUNNING);
	678	break;
	679	case NETDEV_CHANGE:
	680	case NETDEV_GOING_DOWN:
	681	break;
	682	default:
	683	rtmsg_ifinfo(RTM_NEWLINK, dev, 0);
	684	break;
	685	}
	686	return NOTIFY_DONE;
	687	}
	688
	689	static struct notifier_block rtnetlink_dev_notifier = {
	690	.notifier_call = rtnetlink_event,
	691	};
	692
	693	void __init rtnetlink_init(void)
	694	{
	695	int i;
	696
	697	rtattr_max = 0;
	698	for (i = 0; i < ARRAY_SIZE(rta_max); i++)
	699	if (rta_max[i] > rtattr_max)
	700	rtattr_max = rta_max[i];
	701	rta_buf = kmalloc(rtattr_max * sizeof(struct rtattr *), GFP_KERNEL);
	702	if (!rta_buf)
	703	panic("rtnetlink_init: cannot allocate rta_buf\n");
	704
06628607 PM	705	rtnl = netlink_kernel_create(NETLINK_ROUTE, RTNLGRP_MAX, rtnetlink_rcv,
06628607 PM	706	THIS_MODULE);
1da177e4 LT	707	if (rtnl == NULL)
	708	panic("rtnetlink_init: cannot initialize rtnetlink\n");
	709	netlink_set_nonroot(NETLINK_ROUTE, NL_NONROOT_RECV);
	710	register_netdevice_notifier(&rtnetlink_dev_notifier);
	711	rtnetlink_links[PF_UNSPEC] = link_rtnetlink_table;
	712	rtnetlink_links[PF_PACKET] = link_rtnetlink_table;
	713	}
	714
	715	EXPORT_SYMBOL(__rta_fill);
	716	EXPORT_SYMBOL(rtattr_strlcpy);
	717	EXPORT_SYMBOL(rtattr_parse);
	718	EXPORT_SYMBOL(rtnetlink_links);
	719	EXPORT_SYMBOL(rtnetlink_put_metrics);
	720	EXPORT_SYMBOL(rtnl);
	721	EXPORT_SYMBOL(rtnl_lock);
	722	EXPORT_SYMBOL(rtnl_lock_interruptible);
	723	EXPORT_SYMBOL(rtnl_sem);
	724	EXPORT_SYMBOL(rtnl_unlock);