devinet.c

Linux内核源代码 为压缩文件 是<<Linux内核>>一书中的源代码

/*
 *	NET3	IP device support routines.
 *
 *	Version: $Id: devinet.c,v 1.39 2000/12/10 22:24:11 davem Exp $
 *
 *		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.
 *
 *	Derived from the IP parts of dev.c 1.0.19
 * 		Authors:	Ross Biro, <[e-mail]>
 *				Fred N. van Kempen, <[e-mail]>
 *				Mark Evans, <[e-mail]>
 *
 *	Additional Authors:
 *		Alan Cox, <[e-mail]>
 *		Alexey Kuznetsov, <[e-mail]>
 *
 *	Changes:
 *	        Alexey Kuznetsov:	pa_* fields are replaced with ifaddr lists.
 *		Cyrus Durgin:		updated for kmod
 */

#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/string.h>
#include <linux/mm.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/in.h>
#include <linux/errno.h>
#include <linux/interrupt.h>
#include <linux/if_ether.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/skbuff.h>
#include <linux/rtnetlink.h>
#include <linux/init.h>
#include <linux/notifier.h>
#include <linux/inetdevice.h>
#include <linux/igmp.h>
#ifdef CONFIG_SYSCTL
#include <linux/sysctl.h>
#endif
#ifdef CONFIG_KMOD
#include <linux/kmod.h>
#endif

#include <net/ip.h>
#include <net/route.h>
#include <net/ip_fib.h>

struct ipv4_devconf ipv4_devconf = { 1, 1, 1, 1, 0, };
static struct ipv4_devconf ipv4_devconf_dflt = { 1, 1, 1, 1, 1, };

#ifdef CONFIG_RTNETLINK
static void rtmsg_ifa(int event, struct in_ifaddr *);
#else
#define rtmsg_ifa(a,b)	do { } while(0)
#endif

static struct notifier_block *inetaddr_chain;
static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy);
#ifdef CONFIG_SYSCTL
static void devinet_sysctl_register(struct in_device *in_dev, struct ipv4_devconf *p);
static void devinet_sysctl_unregister(struct ipv4_devconf *p);
#endif

int inet_ifa_count;
int inet_dev_count;

/* Locks all the inet devices. */

rwlock_t inetdev_lock = RW_LOCK_UNLOCKED;


static struct in_ifaddr * inet_alloc_ifa(void)
{
	struct in_ifaddr *ifa;

	ifa = kmalloc(sizeof(*ifa), GFP_KERNEL);
	if (ifa) {
		memset(ifa, 0, sizeof(*ifa));
		inet_ifa_count++;
	}

	return ifa;
}

static __inline__ void inet_free_ifa(struct in_ifaddr *ifa)
{
	if (ifa->ifa_dev)
		__in_dev_put(ifa->ifa_dev);
	kfree(ifa);
	inet_ifa_count--;
}

void in_dev_finish_destroy(struct in_device *idev)
{
	struct net_device *dev = idev->dev;

	BUG_TRAP(idev->ifa_list==NULL);
	BUG_TRAP(idev->mc_list==NULL);
#ifdef NET_REFCNT_DEBUG
	printk(KERN_DEBUG "in_dev_finish_destroy: %p=%s\n", idev, dev ? dev->name : "NIL");
#endif
	dev_put(dev);
	if (!idev->dead) {
		printk("Freeing alive in_device %p\n", idev);
		return;
	}
	inet_dev_count--;
	kfree(idev);
}

struct in_device *inetdev_init(struct net_device *dev)
{
	struct in_device *in_dev;

	ASSERT_RTNL();

	in_dev = kmalloc(sizeof(*in_dev), GFP_KERNEL);
	if (!in_dev)
		return NULL;
	memset(in_dev, 0, sizeof(*in_dev));
	in_dev->lock = RW_LOCK_UNLOCKED;
	memcpy(&in_dev->cnf, &ipv4_devconf_dflt, sizeof(in_dev->cnf));
	in_dev->cnf.sysctl = NULL;
	in_dev->dev = dev;
	if ((in_dev->arp_parms = neigh_parms_alloc(dev, &arp_tbl)) == NULL) {
		kfree(in_dev);
		return NULL;
	}
	inet_dev_count++;
	/* Reference in_dev->dev */
	dev_hold(dev);
#ifdef CONFIG_SYSCTL
	neigh_sysctl_register(dev, in_dev->arp_parms, NET_IPV4, NET_IPV4_NEIGH, "ipv4");
#endif
	write_lock_bh(&inetdev_lock);
	dev->ip_ptr = in_dev;
	/* Account for reference dev->ip_ptr */
	in_dev_hold(in_dev);
	write_unlock_bh(&inetdev_lock);
#ifdef CONFIG_SYSCTL
	devinet_sysctl_register(in_dev, &in_dev->cnf);
#endif
	if (dev->flags&IFF_UP)
		ip_mc_up(in_dev);
	return in_dev;
}

static void inetdev_destroy(struct in_device *in_dev)
{
	struct in_ifaddr *ifa;

	ASSERT_RTNL();

	in_dev->dead = 1;

	ip_mc_destroy_dev(in_dev);

	while ((ifa = in_dev->ifa_list) != NULL) {
		inet_del_ifa(in_dev, &in_dev->ifa_list, 0);
		inet_free_ifa(ifa);
	}

#ifdef CONFIG_SYSCTL
	devinet_sysctl_unregister(&in_dev->cnf);
#endif
	write_lock_bh(&inetdev_lock);
	in_dev->dev->ip_ptr = NULL;
	/* in_dev_put following below will kill the in_device */
	write_unlock_bh(&inetdev_lock);


	neigh_parms_release(&arp_tbl, in_dev->arp_parms);
	in_dev_put(in_dev);
}

int inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b)
{
	read_lock(&in_dev->lock);
	for_primary_ifa(in_dev) {
		if (inet_ifa_match(a, ifa)) {
			if (!b || inet_ifa_match(b, ifa)) {
				read_unlock(&in_dev->lock);
				return 1;
			}
		}
	} endfor_ifa(in_dev);
	read_unlock(&in_dev->lock);
	return 0;
} 

static void
inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy)
{
	struct in_ifaddr *ifa1 = *ifap;

	ASSERT_RTNL();

	/* 1. Deleting primary ifaddr forces deletion all secondaries */

	if (!(ifa1->ifa_flags&IFA_F_SECONDARY)) {
		struct in_ifaddr *ifa;
		struct in_ifaddr **ifap1 = &ifa1->ifa_next;

		while ((ifa=*ifap1) != NULL) {
			if (!(ifa->ifa_flags&IFA_F_SECONDARY) ||
			    ifa1->ifa_mask != ifa->ifa_mask ||
			    !inet_ifa_match(ifa1->ifa_address, ifa)) {
				ifap1 = &ifa->ifa_next;
				continue;
			}
			write_lock_bh(&in_dev->lock);
			*ifap1 = ifa->ifa_next;
			write_unlock_bh(&in_dev->lock);

			rtmsg_ifa(RTM_DELADDR, ifa);
			notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa);
			inet_free_ifa(ifa);
		}
	}

	/* 2. Unlink it */

	write_lock_bh(&in_dev->lock);
	*ifap = ifa1->ifa_next;
	write_unlock_bh(&in_dev->lock);

	/* 3. Announce address deletion */

	/* Send message first, then call notifier.
	   At first sight, FIB update triggered by notifier
	   will refer to already deleted ifaddr, that could confuse
	   netlink listeners. It is not true: look, gated sees
	   that route deleted and if it still thinks that ifaddr
	   is valid, it will try to restore deleted routes... Grr.
	   So that, this order is correct.
	 */
	rtmsg_ifa(RTM_DELADDR, ifa1);
	notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa1);
	if (destroy) {
		inet_free_ifa(ifa1);

		if (in_dev->ifa_list == NULL)
			inetdev_destroy(in_dev);
	}
}

static int
inet_insert_ifa(struct in_ifaddr *ifa)
{
	struct in_device *in_dev = ifa->ifa_dev;
	struct in_ifaddr *ifa1, **ifap, **last_primary;

	ASSERT_RTNL();

	if (ifa->ifa_local == 0) {
		inet_free_ifa(ifa);
		return 0;
	}

	ifa->ifa_flags &= ~IFA_F_SECONDARY;
	last_primary = &in_dev->ifa_list;

	for (ifap=&in_dev->ifa_list; (ifa1=*ifap)!=NULL; ifap=&ifa1->ifa_next) {
		if (!(ifa1->ifa_flags&IFA_F_SECONDARY) && ifa->ifa_scope <= ifa1->ifa_scope)
			last_primary = &ifa1->ifa_next;
		if (ifa1->ifa_mask == ifa->ifa_mask && inet_ifa_match(ifa1->ifa_address, ifa)) {
			if (ifa1->ifa_local == ifa->ifa_local) {
				inet_free_ifa(ifa);
				return -EEXIST;
			}
			if (ifa1->ifa_scope != ifa->ifa_scope) {
				inet_free_ifa(ifa);
				return -EINVAL;
			}
			ifa->ifa_flags |= IFA_F_SECONDARY;
		}
	}

	if (!(ifa->ifa_flags&IFA_F_SECONDARY)) {
		net_srandom(ifa->ifa_local);
		ifap = last_primary;
	}

	ifa->ifa_next = *ifap;
	write_lock_bh(&in_dev->lock);
	*ifap = ifa;
	write_unlock_bh(&in_dev->lock);

	/* Send message first, then call notifier.
	   Notifier will trigger FIB update, so that
	   listeners of netlink will know about new ifaddr */
	rtmsg_ifa(RTM_NEWADDR, ifa);
	notifier_call_chain(&inetaddr_chain, NETDEV_UP, ifa);

	return 0;
}

static int
inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa)
{
	struct in_device *in_dev = __in_dev_get(dev);

	ASSERT_RTNL();

	if (in_dev == NULL) {
		in_dev = inetdev_init(dev);
		if (in_dev == NULL) {
			inet_free_ifa(ifa);
			return -ENOBUFS;
		}
	}
	if (ifa->ifa_dev != in_dev) {
		BUG_TRAP(ifa->ifa_dev==NULL);
		in_dev_hold(in_dev);
		ifa->ifa_dev=in_dev;
	}
	if (LOOPBACK(ifa->ifa_local))
		ifa->ifa_scope = RT_SCOPE_HOST;
	return inet_insert_ifa(ifa);
}

struct in_device *inetdev_by_index(int ifindex)
{
	struct net_device *dev;
	struct in_device *in_dev = NULL;
	read_lock(&dev_base_lock);
	dev = __dev_get_by_index(ifindex);
	if (dev)
		in_dev = in_dev_get(dev);
	read_unlock(&dev_base_lock);
	return in_dev;
}

/* Called only from RTNL semaphored context. No locks. */

struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, u32 prefix, u32 mask)
{
	ASSERT_RTNL();

	for_primary_ifa(in_dev) {
		if (ifa->ifa_mask == mask && inet_ifa_match(prefix, ifa))
			return ifa;
	} endfor_ifa(in_dev);
	return NULL;
}

#ifdef CONFIG_RTNETLINK

int
inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
	struct rtattr  **rta = arg;
	struct in_device *in_dev;
	struct ifaddrmsg *ifm = NLMSG_DATA(nlh);
	struct in_ifaddr *ifa, **ifap;

	ASSERT_RTNL();

	if ((in_dev = inetdev_by_index(ifm->ifa_index)) == NULL)
		return -EADDRNOTAVAIL;
	__in_dev_put(in_dev);

	for (ifap=&in_dev->ifa_list; (ifa=*ifap)!=NULL; ifap=&ifa->ifa_next) {
		if ((rta[IFA_LOCAL-1] && memcmp(RTA_DATA(rta[IFA_LOCAL-1]), &ifa->ifa_local, 4)) ||
		    (rta[IFA_LABEL-1] && strcmp(RTA_DATA(rta[IFA_LABEL-1]), ifa->ifa_label)) ||
		    (rta[IFA_ADDRESS-1] &&
		     (ifm->ifa_prefixlen != ifa->ifa_prefixlen ||
		      !inet_ifa_match(*(u32*)RTA_DATA(rta[IFA_ADDRESS-1]), ifa))))
			continue;
		inet_del_ifa(in_dev, ifap, 1);
		return 0;
	}

	return -EADDRNOTAVAIL;
}

int
inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg)
{
	struct rtattr **rta = arg;
	struct net_device *dev;
	struct in_device *in_dev;
	struct ifaddrmsg *ifm = NLMSG_DATA(nlh);
	struct in_ifaddr *ifa;

	ASSERT_RTNL();

	if (ifm->ifa_prefixlen > 32 || rta[IFA_LOCAL-1] == NULL)
		return -EINVAL;

	if ((dev = __dev_get_by_index(ifm->ifa_index)) == NULL)
		return -ENODEV;

	if ((in_dev = __in_dev_get(dev)) == NULL) {
		in_dev = inetdev_init(dev);
		if (!in_dev)
			return -ENOBUFS;
	}

	if ((ifa = inet_alloc_ifa()) == NULL)
		return -ENOBUFS;

	if (rta[IFA_ADDRESS-1] == NULL)
		rta[IFA_ADDRESS-1] = rta[IFA_LOCAL-1];
	memcpy(&ifa->ifa_local, RTA_DATA(rta[IFA_LOCAL-1]), 4);
	memcpy(&ifa->ifa_address, RTA_DATA(rta[IFA_ADDRESS-1]), 4);
	ifa->ifa_prefixlen = ifm->ifa_prefixlen;
	ifa->ifa_mask = inet_make_mask(ifm->ifa_prefixlen);
	if (rta[IFA_BROADCAST-1])
		memcpy(&ifa->ifa_broadcast, RTA_DATA(rta[IFA_BROADCAST-1]), 4);
	if (rta[IFA_ANYCAST-1])
		memcpy(&ifa->ifa_anycast, RTA_DATA(rta[IFA_ANYCAST-1]), 4);
	ifa->ifa_flags = ifm->ifa_flags;
	ifa->ifa_scope = ifm->ifa_scope;
	in_dev_hold(in_dev);
	ifa->ifa_dev = in_dev;
	if (rta[IFA_LABEL-1])
		memcpy(ifa->ifa_label, RTA_DATA(rta[IFA_LABEL-1]), IFNAMSIZ);
	else
		memcpy(ifa->ifa_label, dev->name, IFNAMSIZ);

	return inet_insert_ifa(ifa);
}

#endif

/* 
 *	Determine a default network mask, based on the IP address. 
 */

static __inline__ int inet_abc_len(u32 addr)
{
  	if (ZERONET(addr))
  		return 0;

  	addr = ntohl(addr);
  	if (IN_CLASSA(addr)) 
  		return 8;
  	if (IN_CLASSB(addr)) 
  		return 16;
  	if (IN_CLASSC(addr)) 
  		return 24;

	/*
	 *	Something else, probably a multicast. 
	 */
  	 
  	return -1;
}


int devinet_ioctl(unsigned int cmd, void *arg)
{
	struct ifreq ifr;
	struct sockaddr_in *sin = (struct sockaddr_in *)&ifr.ifr_addr;
	struct in_device *in_dev;
	struct in_ifaddr **ifap = NULL;
	struct in_ifaddr *ifa = NULL;
	struct net_device *dev;
	char *colon;
	int ret = 0;

	/*
	 *	Fetch the caller's info block into kernel space
	 */

	if (copy_from_user(&ifr, arg, sizeof(struct ifreq)))
		return -EFAULT;
	ifr.ifr_name[IFNAMSIZ-1] = 0;

	colon = strchr(ifr.ifr_name, ':');
	if (colon)
		*colon = 0;

#ifdef CONFIG_KMOD
	dev_load(ifr.ifr_name);
#endif

	switch(cmd) {
	case SIOCGIFADDR:	/* Get interface address */
	case SIOCGIFBRDADDR:	/* Get the broadcast address */
	case SIOCGIFDSTADDR:	/* Get the destination address */
	case SIOCGIFNETMASK:	/* Get the netmask for the interface */
		/* Note that this ioctls will not sleep,
		   so that we do not impose a lock.
		   One day we will be forced to put shlock here (I mean SMP)
		 */
		memset(sin, 0, sizeof(*sin));
		sin->sin_family = AF_INET;
		break;

	case SIOCSIFFLAGS:
		if (!capable(CAP_NET_ADMIN))
			return -EACCES;
		break;
	case SIOCSIFADDR:	/* Set interface address (and family) */
	case SIOCSIFBRDADDR:	/* Set the broadcast address */
	case SIOCSIFDSTADDR:	/* Set the destination address */
	case SIOCSIFNETMASK: 	/* Set the netmask for the interface */
		if (!capable(CAP_NET_ADMIN))
			return -EACCES;
		if (sin->sin_family != AF_INET)
			return -EINVAL;
		break;
	default:
		return -EINVAL;
	}

	dev_probe_lock();
	rtnl_lock();

	if ((dev = __dev_get_by_name(ifr.ifr_name)) == NULL) {
		ret = -ENODEV;
		goto done;
	}

	if (colon)
		*colon = ':';

	if ((in_dev=__in_dev_get(dev)) != NULL) {
		for (ifap=&in_dev->ifa_list; (ifa=*ifap) != NULL; ifap=&ifa->ifa_next)
			if (strcmp(ifr.ifr_name, ifa->ifa_label) == 0)
				break;
	}

	if (ifa == NULL && cmd != SIOCSIFADDR && cmd != SIOCSIFFLAGS) {
		ret = -EADDRNOTAVAIL;
		goto done;
	}

	switch(cmd) {
		case SIOCGIFADDR:	/* Get interface address */
			sin->sin_addr.s_addr = ifa->ifa_local;
			goto rarok;

		case SIOCGIFBRDADDR:	/* Get the broadcast address */
			sin->sin_addr.s_addr = ifa->ifa_broadcast;
			goto rarok;

		case SIOCGIFDSTADDR:	/* Get the destination address */
			sin->sin_addr.s_addr = ifa->ifa_address;
			goto rarok;

		case SIOCGIFNETMASK:	/* Get the netmask for the interface */
			sin->sin_addr.s_addr = ifa->ifa_mask;
			goto rarok;

		case SIOCSIFFLAGS:
			if (colon) {
				if (ifa == NULL) {
					ret = -EADDRNOTAVAIL;
					break;
				}
				if (!(ifr.ifr_flags&IFF_UP))
					inet_del_ifa(in_dev, ifap, 1);