[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>
#include <net/netlink.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 void set_operstate(struct net_device *dev, unsigned char transition)
{
	unsigned char operstate = dev->operstate;

	switch(transition) {
	case IF_OPER_UP:
		if ((operstate == IF_OPER_DORMANT ||
		     operstate == IF_OPER_UNKNOWN) &&
		    !netif_dormant(dev))
			operstate = IF_OPER_UP;
		break;

	case IF_OPER_DORMANT:
		if (operstate == IF_OPER_UP ||
		    operstate == IF_OPER_UNKNOWN)
			operstate = IF_OPER_DORMANT;
		break;
	};

	if (dev->operstate != operstate) {
		write_lock_bh(&dev_base_lock);
		dev->operstate = operstate;
		write_unlock_bh(&dev_base_lock);
		netdev_state_change(dev);
	}
}

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) {
		u8 operstate = netif_running(dev)?dev->operstate:IF_OPER_DOWN;
		u8 link_mode = dev->link_mode;
		RTA_PUT(skb, IFLA_OPERSTATE, sizeof(operstate), &operstate);
		RTA_PUT(skb, IFLA_LINKMODE, sizeof(link_mode), &link_mode);
	}

	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 (ida[IFLA_OPERSTATE - 1]) {
		if (ida[IFLA_OPERSTATE - 1]->rta_len != RTA_LENGTH(sizeof(u8)))
			goto out;

		set_operstate(dev, *((u8 *) RTA_DATA(ida[IFLA_OPERSTATE - 1])));
	}

	if (ida[IFLA_LINKMODE - 1]) {
		if (ida[IFLA_LINKMODE - 1]->rta_len != RTA_LENGTH(sizeof(u8)))
			goto out;

		write_lock_bh(&dev_base_lock);
		dev->link_mode = *((u8 *) RTA_DATA(ida[IFLA_LINKMODE - 1]));
		write_unlock_bh(&dev_base_lock);
	}

	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, 0, 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) {
		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;
		}

		netlink_queue_skip(nlh, skb);
		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;
}

static void rtnetlink_rcv(struct sock *sk, int len)
{
	unsigned int qlen = 0;

	do {
		rtnl_lock();
		netlink_run_queue(sk, &qlen, &rtnetlink_rcv_msg);
		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>
9ac4a169	52	#include <net/netlink.h>
1da177e4 LT	53
	54	DECLARE_MUTEX(rtnl_sem);
	55
	56	void rtnl_lock(void)
	57	{
	58	rtnl_shlock();
	59	}
	60
	61	int rtnl_lock_interruptible(void)
	62	{
	63	return down_interruptible(&rtnl_sem);
	64	}
	65
	66	void rtnl_unlock(void)
	67	{
	68	rtnl_shunlock();
	69
	70	netdev_run_todo();
	71	}
	72
	73	int rtattr_parse(struct rtattr tb[], int maxattr, struct rtattr rta, int len)
	74	{
	75	memset(tb, 0, sizeof(struct rtattr)maxattr);
	76
	77	while (RTA_OK(rta, len)) {
	78	unsigned flavor = rta->rta_type;
	79	if (flavor && flavor <= maxattr)
	80	tb[flavor-1] = rta;
	81	rta = RTA_NEXT(rta, len);
	82	}
	83	return 0;
	84	}
	85
	86	struct sock *rtnl;
	87
	88	struct rtnetlink_link * rtnetlink_links[NPROTO];
	89
db46edc6	90	static const int rtm_min[RTM_NR_FAMILIES] =
1da177e4	91	{
f90a0a74 TG	92	[RTM_FAM(RTM_NEWLINK)] = NLMSG_LENGTH(sizeof(struct ifinfomsg)),
	93	[RTM_FAM(RTM_NEWADDR)] = NLMSG_LENGTH(sizeof(struct ifaddrmsg)),
	94	[RTM_FAM(RTM_NEWROUTE)] = NLMSG_LENGTH(sizeof(struct rtmsg)),
	95	[RTM_FAM(RTM_NEWNEIGH)] = NLMSG_LENGTH(sizeof(struct ndmsg)),
	96	[RTM_FAM(RTM_NEWRULE)] = NLMSG_LENGTH(sizeof(struct rtmsg)),
	97	[RTM_FAM(RTM_NEWQDISC)] = NLMSG_LENGTH(sizeof(struct tcmsg)),
	98	[RTM_FAM(RTM_NEWTCLASS)] = NLMSG_LENGTH(sizeof(struct tcmsg)),
	99	[RTM_FAM(RTM_NEWTFILTER)] = NLMSG_LENGTH(sizeof(struct tcmsg)),
	100	[RTM_FAM(RTM_NEWACTION)] = NLMSG_LENGTH(sizeof(struct tcamsg)),
	101	[RTM_FAM(RTM_NEWPREFIX)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	102	[RTM_FAM(RTM_GETMULTICAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
	103	[RTM_FAM(RTM_GETANYCAST)] = NLMSG_LENGTH(sizeof(struct rtgenmsg)),
c7fb64db	104	[RTM_FAM(RTM_NEWNEIGHTBL)] = NLMSG_LENGTH(sizeof(struct ndtmsg)),
1da177e4 LT	105	};
1da177e4 LT	106
db46edc6	107	static const int rta_max[RTM_NR_FAMILIES] =
1da177e4	108	{
f90a0a74 TG	109	[RTM_FAM(RTM_NEWLINK)] = IFLA_MAX,
	110	[RTM_FAM(RTM_NEWADDR)] = IFA_MAX,
	111	[RTM_FAM(RTM_NEWROUTE)] = RTA_MAX,
	112	[RTM_FAM(RTM_NEWNEIGH)] = NDA_MAX,
	113	[RTM_FAM(RTM_NEWRULE)] = RTA_MAX,
	114	[RTM_FAM(RTM_NEWQDISC)] = TCA_MAX,
	115	[RTM_FAM(RTM_NEWTCLASS)] = TCA_MAX,
	116	[RTM_FAM(RTM_NEWTFILTER)] = TCA_MAX,
	117	[RTM_FAM(RTM_NEWACTION)] = TCAA_MAX,
c7fb64db	118	[RTM_FAM(RTM_NEWNEIGHTBL)] = NDTA_MAX,
1da177e4 LT	119	};
	120
	121	void __rta_fill(struct sk_buff skb, int attrtype, int attrlen, const void data)
	122	{
	123	struct rtattr *rta;
	124	int size = RTA_LENGTH(attrlen);
	125
	126	rta = (struct rtattr*)skb_put(skb, RTA_ALIGN(size));
	127	rta->rta_type = attrtype;
	128	rta->rta_len = size;
	129	memcpy(RTA_DATA(rta), data, attrlen);
b3563c4f	130	memset(RTA_DATA(rta) + attrlen, 0, RTA_ALIGN(size) - size);
1da177e4 LT	131	}
	132
	133	size_t rtattr_strlcpy(char dest, const struct rtattr rta, size_t size)
	134	{
	135	size_t ret = RTA_PAYLOAD(rta);
	136	char *src = RTA_DATA(rta);
	137
	138	if (ret > 0 && src[ret - 1] == '\0')
	139	ret--;
	140	if (size > 0) {
	141	size_t len = (ret >= size) ? size - 1 : ret;
	142	memset(dest, 0, size);
	143	memcpy(dest, src, len);
	144	}
	145	return ret;
	146	}
	147
	148	int rtnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo)
	149	{
	150	int err = 0;
	151
ac6d439d	152	NETLINK_CB(skb).dst_group = group;
1da177e4 LT	153	if (echo)
	154	atomic_inc(&skb->users);
	155	netlink_broadcast(rtnl, skb, pid, group, GFP_KERNEL);
	156	if (echo)
	157	err = netlink_unicast(rtnl, skb, pid, MSG_DONTWAIT);
	158	return err;
	159	}
	160
	161	int rtnetlink_put_metrics(struct sk_buff skb, u32 metrics)
	162	{
	163	struct rtattr mx = (struct rtattr)skb->tail;
	164	int i;
	165
	166	RTA_PUT(skb, RTA_METRICS, 0, NULL);
	167	for (i=0; i<RTAX_MAX; i++) {
	168	if (metrics[i])
	169	RTA_PUT(skb, i+1, sizeof(u32), metrics+i);
	170	}
	171	mx->rta_len = skb->tail - (u8*)mx;
	172	if (mx->rta_len == RTA_LENGTH(0))
	173	skb_trim(skb, (u8*)mx - skb->data);
	174	return 0;
	175
	176	rtattr_failure:
	177	skb_trim(skb, (u8*)mx - skb->data);
	178	return -1;
	179	}
	180
	181
b00055aa SR	182	static void set_operstate(struct net_device *dev, unsigned char transition)
	183	{
	184	unsigned char operstate = dev->operstate;
	185
	186	switch(transition) {
	187	case IF_OPER_UP:
	188	if ((operstate == IF_OPER_DORMANT \|\|
	189	operstate == IF_OPER_UNKNOWN) &&
	190	!netif_dormant(dev))
	191	operstate = IF_OPER_UP;
	192	break;
	193
	194	case IF_OPER_DORMANT:
	195	if (operstate == IF_OPER_UP \|\|
	196	operstate == IF_OPER_UNKNOWN)
	197	operstate = IF_OPER_DORMANT;
	198	break;
	199	};
	200
	201	if (dev->operstate != operstate) {
	202	write_lock_bh(&dev_base_lock);
	203	dev->operstate = operstate;
	204	write_unlock_bh(&dev_base_lock);
	205	netdev_state_change(dev);
	206	}
	207	}
	208
1da177e4	209	static int rtnetlink_fill_ifinfo(struct sk_buff skb, struct net_device dev,
b6544c0b JHS	210	int type, u32 pid, u32 seq, u32 change,
b6544c0b JHS	211	unsigned int flags)
1da177e4 LT	212	{
	213	struct ifinfomsg *r;
	214	struct nlmsghdr *nlh;
	215	unsigned char *b = skb->tail;
	216
b6544c0b	217	nlh = NLMSG_NEW(skb, pid, seq, type, sizeof(*r), flags);
1da177e4 LT	218	r = NLMSG_DATA(nlh);
1da177e4 LT	219	r->ifi_family = AF_UNSPEC;
9ef1d4c7	220	r->__ifi_pad = 0;
1da177e4 LT	221	r->ifi_type = dev->type;
	222	r->ifi_index = dev->ifindex;
	223	r->ifi_flags = dev_get_flags(dev);
	224	r->ifi_change = change;
	225
	226	RTA_PUT(skb, IFLA_IFNAME, strlen(dev->name)+1, dev->name);
	227
	228	if (1) {
	229	u32 txqlen = dev->tx_queue_len;
	230	RTA_PUT(skb, IFLA_TXQLEN, sizeof(txqlen), &txqlen);
	231	}
	232
	233	if (1) {
	234	u32 weight = dev->weight;
	235	RTA_PUT(skb, IFLA_WEIGHT, sizeof(weight), &weight);
	236	}
	237
b00055aa SR	238	if (1) {
	239	u8 operstate = netif_running(dev)?dev->operstate:IF_OPER_DOWN;
	240	u8 link_mode = dev->link_mode;
	241	RTA_PUT(skb, IFLA_OPERSTATE, sizeof(operstate), &operstate);
	242	RTA_PUT(skb, IFLA_LINKMODE, sizeof(link_mode), &link_mode);
	243	}
	244
1da177e4 LT	245	if (1) {
	246	struct rtnl_link_ifmap map = {
	247	.mem_start = dev->mem_start,
	248	.mem_end = dev->mem_end,
	249	.base_addr = dev->base_addr,
	250	.irq = dev->irq,
	251	.dma = dev->dma,
	252	.port = dev->if_port,
	253	};
	254	RTA_PUT(skb, IFLA_MAP, sizeof(map), &map);
	255	}
	256
	257	if (dev->addr_len) {
	258	RTA_PUT(skb, IFLA_ADDRESS, dev->addr_len, dev->dev_addr);
	259	RTA_PUT(skb, IFLA_BROADCAST, dev->addr_len, dev->broadcast);
	260	}
	261
	262	if (1) {
	263	u32 mtu = dev->mtu;
	264	RTA_PUT(skb, IFLA_MTU, sizeof(mtu), &mtu);
	265	}
	266
	267	if (dev->ifindex != dev->iflink) {
	268	u32 iflink = dev->iflink;
	269	RTA_PUT(skb, IFLA_LINK, sizeof(iflink), &iflink);
	270	}
	271
	272	if (dev->qdisc_sleeping)
	273	RTA_PUT(skb, IFLA_QDISC,
	274	strlen(dev->qdisc_sleeping->ops->id) + 1,
	275	dev->qdisc_sleeping->ops->id);
	276
	277	if (dev->master) {
	278	u32 master = dev->master->ifindex;
	279	RTA_PUT(skb, IFLA_MASTER, sizeof(master), &master);
	280	}
	281
	282	if (dev->get_stats) {
	283	unsigned long stats = (unsigned long)dev->get_stats(dev);
	284	if (stats) {
	285	struct rtattr *a;
	286	__u32 *s;
	287	int i;
	288	int n = sizeof(struct rtnl_link_stats)/4;
	289
	290	a = __RTA_PUT(skb, IFLA_STATS, n*4);
	291	s = RTA_DATA(a);
	292	for (i=0; i<n; i++)
	293	s[i] = stats[i];
	294	}
	295	}
	296	nlh->nlmsg_len = skb->tail - b;
	297	return skb->len;
	298
	299	nlmsg_failure:
	300	rtattr_failure:
	301	skb_trim(skb, b - skb->data);
	302	return -1;
	303	}
	304
	305	static int rtnetlink_dump_ifinfo(struct sk_buff skb, struct netlink_callback cb)
	306	{
	307	int idx;
	308	int s_idx = cb->args[0];
309	struct net_device *dev;
310
311	read_lock(&dev_base_lock);
312	for (dev=dev_base, idx=0; dev; dev = dev->next, idx++) {
313	if (idx < s_idx)
314	continue;
b6544c0b JHS	315	if (rtnetlink_fill_ifinfo(skb, dev, RTM_NEWLINK,
	316	NETLINK_CB(cb->skb).pid,
	317	cb->nlh->nlmsg_seq, 0,
	318	NLM_F_MULTI) <= 0)
1da177e4 LT	319	break;
	320	}
	321	read_unlock(&dev_base_lock);
	322	cb->args[0] = idx;
	323
	324	return skb->len;
	325	}
	326
	327	static int do_setlink(struct sk_buff skb, struct nlmsghdr nlh, void *arg)
	328	{
	329	struct ifinfomsg *ifm = NLMSG_DATA(nlh);
	330	struct rtattr **ida = arg;
	331	struct net_device *dev;
	332	int err, send_addr_notify = 0;
	333
	334	if (ifm->ifi_index >= 0)
	335	dev = dev_get_by_index(ifm->ifi_index);
	336	else if (ida[IFLA_IFNAME - 1]) {
	337	char ifname[IFNAMSIZ];
	338
	339	if (rtattr_strlcpy(ifname, ida[IFLA_IFNAME - 1],
	340	IFNAMSIZ) >= IFNAMSIZ)
	341	return -EINVAL;
	342	dev = dev_get_by_name(ifname);
	343	} else
	344	return -EINVAL;
	345
	346	if (!dev)
	347	return -ENODEV;
	348
	349	err = -EINVAL;
	350
	351	if (ifm->ifi_flags)
	352	dev_change_flags(dev, ifm->ifi_flags);
	353
	354	if (ida[IFLA_MAP - 1]) {
	355	struct rtnl_link_ifmap *u_map;
	356	struct ifmap k_map;
	357
	358	if (!dev->set_config) {
	359	err = -EOPNOTSUPP;
	360	goto out;
	361	}
	362
	363	if (!netif_device_present(dev)) {
	364	err = -ENODEV;
	365	goto out;
	366	}
	367
	368	if (ida[IFLA_MAP - 1]->rta_len != RTA_LENGTH(sizeof(*u_map)))
	369	goto out;
	370
	371	u_map = RTA_DATA(ida[IFLA_MAP - 1]);
	372
	373	k_map.mem_start = (unsigned long) u_map->mem_start;
	374	k_map.mem_end = (unsigned long) u_map->mem_end;
	375	k_map.base_addr = (unsigned short) u_map->base_addr;
	376	k_map.irq = (unsigned char) u_map->irq;
	377	k_map.dma = (unsigned char) u_map->dma;
	378	k_map.port = (unsigned char) u_map->port;
	379
	380	err = dev->set_config(dev, &k_map);
	381
	382	if (err)
383	goto out;
384	}
385
386	if (ida[IFLA_ADDRESS - 1]) {
387	if (!dev->set_mac_address) {
388	err = -EOPNOTSUPP;
389	goto out;
390	}
391	if (!netif_device_present(dev)) {
392	err = -ENODEV;
393	goto out;
394	}
395	if (ida[IFLA_ADDRESS - 1]->rta_len != RTA_LENGTH(dev->addr_len))
396	goto out;
397
398	err = dev->set_mac_address(dev, RTA_DATA(ida[IFLA_ADDRESS - 1]));
399	if (err)
400	goto out;
401	send_addr_notify = 1;
402	}
403
404	if (ida[IFLA_BROADCAST - 1]) {
405	if (ida[IFLA_BROADCAST - 1]->rta_len != RTA_LENGTH(dev->addr_len))
406	goto out;
407	memcpy(dev->broadcast, RTA_DATA(ida[IFLA_BROADCAST - 1]),
408	dev->addr_len);
409	send_addr_notify = 1;
410	}
411
412	if (ida[IFLA_MTU - 1]) {
413	if (ida[IFLA_MTU - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
414	goto out;
415	err = dev_set_mtu(dev, ((u32 ) RTA_DATA(ida[IFLA_MTU - 1])));
416
417	if (err)
418	goto out;
419
420	}
421
422	if (ida[IFLA_TXQLEN - 1]) {
423	if (ida[IFLA_TXQLEN - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
424	goto out;
425
426	dev->tx_queue_len = ((u32 ) RTA_DATA(ida[IFLA_TXQLEN - 1]));
427	}
428
429	if (ida[IFLA_WEIGHT - 1]) {
430	if (ida[IFLA_WEIGHT - 1]->rta_len != RTA_LENGTH(sizeof(u32)))
431	goto out;
432
433	dev->weight = ((u32 ) RTA_DATA(ida[IFLA_WEIGHT - 1]));
434	}
435
b00055aa SR	436	if (ida[IFLA_OPERSTATE - 1]) {
	437	if (ida[IFLA_OPERSTATE - 1]->rta_len != RTA_LENGTH(sizeof(u8)))
	438	goto out;
	439
	440	set_operstate(dev, ((u8 ) RTA_DATA(ida[IFLA_OPERSTATE - 1])));
	441	}
	442
	443	if (ida[IFLA_LINKMODE - 1]) {
	444	if (ida[IFLA_LINKMODE - 1]->rta_len != RTA_LENGTH(sizeof(u8)))
	445	goto out;
	446
	447	write_lock_bh(&dev_base_lock);
	448	dev->link_mode = ((u8 ) RTA_DATA(ida[IFLA_LINKMODE - 1]));
	449	write_unlock_bh(&dev_base_lock);
	450	}
	451
1da177e4 LT	452	if (ifm->ifi_index >= 0 && ida[IFLA_IFNAME - 1]) {
	453	char ifname[IFNAMSIZ];
	454
	455	if (rtattr_strlcpy(ifname, ida[IFLA_IFNAME - 1],
	456	IFNAMSIZ) >= IFNAMSIZ)
	457	goto out;
	458	err = dev_change_name(dev, ifname);
	459	if (err)
	460	goto out;
	461	}
	462
	463	err = 0;
	464
	465	out:
	466	if (send_addr_notify)
	467	call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
	468
	469	dev_put(dev);
	470	return err;
	471	}
	472
	473	static int rtnetlink_dump_all(struct sk_buff skb, struct netlink_callback cb)
	474	{
	475	int idx;
	476	int s_idx = cb->family;
	477
	478	if (s_idx == 0)
	479	s_idx = 1;
	480	for (idx=1; idx<NPROTO; idx++) {
	481	int type = cb->nlh->nlmsg_type-RTM_BASE;
	482	if (idx < s_idx \|\| idx == PF_PACKET)
	483	continue;
	484	if (rtnetlink_links[idx] == NULL \|\|
	485	rtnetlink_links[idx][type].dumpit == NULL)
	486	continue;
	487	if (idx > s_idx)
	488	memset(&cb->args[0], 0, sizeof(cb->args));
	489	if (rtnetlink_links[idx][type].dumpit(skb, cb))
	490	break;
	491	}
	492	cb->family = idx;
	493
	494	return skb->len;
	495	}
	496
	497	void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
	498	{
	499	struct sk_buff *skb;
	500	int size = NLMSG_SPACE(sizeof(struct ifinfomsg) +
	501	sizeof(struct rtnl_link_ifmap) +
	502	sizeof(struct rtnl_link_stats) + 128);
	503
	504	skb = alloc_skb(size, GFP_KERNEL);
	505	if (!skb)
	506	return;
	507
28633514	508	if (rtnetlink_fill_ifinfo(skb, dev, type, 0, 0, change, 0) < 0) {
1da177e4 LT	509	kfree_skb(skb);
	510	return;
	511	}
ac6d439d PM	512	NETLINK_CB(skb).dst_group = RTNLGRP_LINK;
ac6d439d PM	513	netlink_broadcast(rtnl, skb, 0, RTNLGRP_LINK, GFP_KERNEL);
1da177e4 LT	514	}
1da177e4 LT	515
1da177e4 LT	516	/* Protected by RTNL sempahore. */
	517	static struct rtattr **rta_buf;
	518	static int rtattr_max;
	519
	520	/* Process one rtnetlink message. */
	521
	522	static __inline__ int
	523	rtnetlink_rcv_msg(struct sk_buff skb, struct nlmsghdr nlh, int *errp)
	524	{
	525	struct rtnetlink_link *link;
	526	struct rtnetlink_link *link_tab;
	527	int sz_idx, kind;
	528	int min_len;
	529	int family;
	530	int type;
	531	int err;
	532
	533	/* Only requests are handled by kernel now */
	534	if (!(nlh->nlmsg_flags&NLM_F_REQUEST))
	535	return 0;
	536
	537	type = nlh->nlmsg_type;
	538
	539	/* A control message: ignore them */
	540	if (type < RTM_BASE)
	541	return 0;
	542
	543	/* Unknown message: reply with EINVAL */
	544	if (type > RTM_MAX)
	545	goto err_inval;
	546
	547	type -= RTM_BASE;
	548
	549	/* All the messages must have at least 1 byte length */
	550	if (nlh->nlmsg_len < NLMSG_LENGTH(sizeof(struct rtgenmsg)))
	551	return 0;
	552
	553	family = ((struct rtgenmsg*)NLMSG_DATA(nlh))->rtgen_family;
	554	if (family >= NPROTO) {
	555	*errp = -EAFNOSUPPORT;
	556	return -1;
	557	}
	558
	559	link_tab = rtnetlink_links[family];
	560	if (link_tab == NULL)
	561	link_tab = rtnetlink_links[PF_UNSPEC];
	562	link = &link_tab[type];
	563
	564	sz_idx = type>>2;
	565	kind = type&3;
	566
	567	if (kind != 2 && security_netlink_recv(skb)) {
	568	*errp = -EPERM;
	569	return -1;
	570	}
	571
	572	if (kind == 2 && nlh->nlmsg_flags&NLM_F_DUMP) {
1da177e4 LT	573	if (link->dumpit == NULL)
	574	link = &(rtnetlink_links[PF_UNSPEC][type]);
	575
	576	if (link->dumpit == NULL)
	577	goto err_inval;
	578
	579	if ((*errp = netlink_dump_start(rtnl, skb, nlh,
a8f74b22	580	link->dumpit, NULL)) != 0) {
1da177e4 LT	581	return -1;
1da177e4 LT	582	}
9ac4a169 TG	583
9ac4a169 TG	584	netlink_queue_skip(nlh, skb);
1da177e4 LT	585	return -1;
	586	}
	587
	588	memset(rta_buf, 0, (rtattr_max * sizeof(struct rtattr *)));
	589
	590	min_len = rtm_min[sz_idx];
	591	if (nlh->nlmsg_len < min_len)
	592	goto err_inval;
	593
	594	if (nlh->nlmsg_len > min_len) {
	595	int attrlen = nlh->nlmsg_len - NLMSG_ALIGN(min_len);
	596	struct rtattr attr = (void)nlh + NLMSG_ALIGN(min_len);
	597
	598	while (RTA_OK(attr, attrlen)) {
	599	unsigned flavor = attr->rta_type;
	600	if (flavor) {
	601	if (flavor > rta_max[sz_idx])
	602	goto err_inval;
	603	rta_buf[flavor-1] = attr;
	604	}
	605	attr = RTA_NEXT(attr, attrlen);
	606	}
	607	}
	608
	609	if (link->doit == NULL)
	610	link = &(rtnetlink_links[PF_UNSPEC][type]);
	611	if (link->doit == NULL)
	612	goto err_inval;
	613	err = link->doit(skb, nlh, (void *)&rta_buf[0]);
	614
	615	*errp = err;
	616	return err;
	617
	618	err_inval:
	619	*errp = -EINVAL;
	620	return -1;
	621	}
	622
1da177e4 LT	623	static void rtnetlink_rcv(struct sock *sk, int len)
1da177e4 LT	624	{
9ac4a169	625	unsigned int qlen = 0;
2a0a6ebe	626
1da177e4	627	do {
2a0a6ebe	628	rtnl_lock();
9ac4a169	629	netlink_run_queue(sk, &qlen, &rtnetlink_rcv_msg);
1da177e4 LT	630	up(&rtnl_sem);
	631
	632	netdev_run_todo();
2a0a6ebe	633	} while (qlen);
1da177e4 LT	634	}
1da177e4 LT	635
db46edc6	636	static struct rtnetlink_link link_rtnetlink_table[RTM_NR_MSGTYPES] =
1da177e4	637	{
c7fb64db TG	638	[RTM_GETLINK - RTM_BASE] = { .dumpit = rtnetlink_dump_ifinfo },
	639	[RTM_SETLINK - RTM_BASE] = { .doit = do_setlink },
	640	[RTM_GETADDR - RTM_BASE] = { .dumpit = rtnetlink_dump_all },
	641	[RTM_GETROUTE - RTM_BASE] = { .dumpit = rtnetlink_dump_all },
	642	[RTM_NEWNEIGH - RTM_BASE] = { .doit = neigh_add },
	643	[RTM_DELNEIGH - RTM_BASE] = { .doit = neigh_delete },
	644	[RTM_GETNEIGH - RTM_BASE] = { .dumpit = neigh_dump_info },
	645	[RTM_GETRULE - RTM_BASE] = { .dumpit = rtnetlink_dump_all },
	646	[RTM_GETNEIGHTBL - RTM_BASE] = { .dumpit = neightbl_dump_info },
	647	[RTM_SETNEIGHTBL - RTM_BASE] = { .doit = neightbl_set },
1da177e4 LT	648	};
	649
	650	static int rtnetlink_event(struct notifier_block this, unsigned long event, void ptr)
	651	{
	652	struct net_device *dev = ptr;
	653	switch (event) {
	654	case NETDEV_UNREGISTER:
	655	rtmsg_ifinfo(RTM_DELLINK, dev, ~0U);
	656	break;
	657	case NETDEV_REGISTER:
	658	rtmsg_ifinfo(RTM_NEWLINK, dev, ~0U);
	659	break;
	660	case NETDEV_UP:
	661	case NETDEV_DOWN:
	662	rtmsg_ifinfo(RTM_NEWLINK, dev, IFF_UP\|IFF_RUNNING);
	663	break;
	664	case NETDEV_CHANGE:
	665	case NETDEV_GOING_DOWN:
	666	break;
	667	default:
	668	rtmsg_ifinfo(RTM_NEWLINK, dev, 0);
	669	break;
	670	}
	671	return NOTIFY_DONE;
	672	}
	673
	674	static struct notifier_block rtnetlink_dev_notifier = {
	675	.notifier_call = rtnetlink_event,
	676	};
	677
	678	void __init rtnetlink_init(void)
	679	{
	680	int i;
	681
	682	rtattr_max = 0;
	683	for (i = 0; i < ARRAY_SIZE(rta_max); i++)
	684	if (rta_max[i] > rtattr_max)
	685	rtattr_max = rta_max[i];
	686	rta_buf = kmalloc(rtattr_max * sizeof(struct rtattr *), GFP_KERNEL);
	687	if (!rta_buf)
	688	panic("rtnetlink_init: cannot allocate rta_buf\n");
	689
06628607 PM	690	rtnl = netlink_kernel_create(NETLINK_ROUTE, RTNLGRP_MAX, rtnetlink_rcv,
06628607 PM	691	THIS_MODULE);
1da177e4 LT	692	if (rtnl == NULL)
	693	panic("rtnetlink_init: cannot initialize rtnetlink\n");
	694	netlink_set_nonroot(NETLINK_ROUTE, NL_NONROOT_RECV);
	695	register_netdevice_notifier(&rtnetlink_dev_notifier);
	696	rtnetlink_links[PF_UNSPEC] = link_rtnetlink_table;
	697	rtnetlink_links[PF_PACKET] = link_rtnetlink_table;
	698	}
	699
	700	EXPORT_SYMBOL(__rta_fill);
	701	EXPORT_SYMBOL(rtattr_strlcpy);
	702	EXPORT_SYMBOL(rtattr_parse);
	703	EXPORT_SYMBOL(rtnetlink_links);
	704	EXPORT_SYMBOL(rtnetlink_put_metrics);
	705	EXPORT_SYMBOL(rtnl);
	706	EXPORT_SYMBOL(rtnl_lock);
	707	EXPORT_SYMBOL(rtnl_lock_interruptible);
	708	EXPORT_SYMBOL(rtnl_sem);
	709	EXPORT_SYMBOL(rtnl_unlock);