devinet.c

Linux Kernel 2.6.9 for OMAP1710

/*
 *	NET3	IP device support routines.
 *
 *	Version: $Id: devinet.c,v 1.44 2001/10/31 21:55:54 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
 *		Matthias Andree:	in devinet_ioctl, compare label and
 *					address (4.4BSD alias style support),
 *					fall back to comparing just the label
 *					if no match found.
 */

#include <linux/config.h>

#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/bitops.h>
#include <linux/module.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
#include <linux/kmod.h>

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

struct ipv4_devconf ipv4_devconf = {
	.accept_redirects = 1,
	.send_redirects =  1,
	.secure_redirects = 1,
	.shared_media =	  1,
};

static struct ipv4_devconf ipv4_devconf_dflt = {
	.accept_redirects =  1,
	.send_redirects =    1,
	.secure_redirects =  1,
	.shared_media =	     1,
	.accept_source_route = 1,
};

static void rtmsg_ifa(int event, struct in_ifaddr *);

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

/* Locks all the inet devices. */

static struct in_ifaddr *inet_alloc_ifa(void)
{
	struct in_ifaddr *ifa = kmalloc(sizeof(*ifa), GFP_KERNEL);

	if (ifa) {
		memset(ifa, 0, sizeof(*ifa));
		INIT_RCU_HEAD(&ifa->rcu_head);
	}

	return ifa;
}

static void inet_rcu_free_ifa(struct rcu_head *head)
{
	struct in_ifaddr *ifa = container_of(head, struct in_ifaddr, rcu_head);
	if (ifa->ifa_dev)
		in_dev_put(ifa->ifa_dev);
	kfree(ifa);
}

static inline void inet_free_ifa(struct in_ifaddr *ifa)
{
	call_rcu(&ifa->rcu_head, inet_rcu_free_ifa);
}

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

	BUG_TRAP(!idev->ifa_list);
	BUG_TRAP(!idev->mc_list);
#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);
	else {
		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)
		goto out;
	memset(in_dev, 0, sizeof(*in_dev));
	INIT_RCU_HEAD(&in_dev->rcu_head);
	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)
		goto out_kfree;
	/* Reference in_dev->dev */
	dev_hold(dev);
#ifdef CONFIG_SYSCTL
	neigh_sysctl_register(dev, in_dev->arp_parms, NET_IPV4,
			      NET_IPV4_NEIGH, "ipv4", NULL);
#endif

	/* Account for reference dev->ip_ptr */
	in_dev_hold(in_dev);
	smp_wmb();
	dev->ip_ptr = in_dev;

#ifdef CONFIG_SYSCTL
	devinet_sysctl_register(in_dev, &in_dev->cnf);
#endif
	ip_mc_init_dev(in_dev);
	if (dev->flags & IFF_UP)
		ip_mc_up(in_dev);
out:
	return in_dev;
out_kfree:
	kfree(in_dev);
	in_dev = NULL;
	goto out;
}

static void in_dev_rcu_put(struct rcu_head *head)
{
	struct in_device *idev = container_of(head, struct in_device, rcu_head);
	in_dev_put(idev);
}

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

	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

	dev = in_dev->dev;
	dev->ip_ptr = NULL;

#ifdef CONFIG_SYSCTL
	neigh_sysctl_unregister(in_dev->arp_parms);
#endif
	neigh_parms_release(&arp_tbl, in_dev->arp_parms);
	arp_ifdown(dev);

	call_rcu(&in_dev->rcu_head, in_dev_rcu_put);
}

int inet_addr_onlink(struct in_device *in_dev, u32 a, u32 b)
{
	rcu_read_lock();
	for_primary_ifa(in_dev) {
		if (inet_ifa_match(a, ifa)) {
			if (!b || inet_ifa_match(b, ifa)) {
				rcu_read_unlock();
				return 1;
			}
		}
	} endfor_ifa(in_dev);
	rcu_read_unlock();
	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;
			}

			*ifap1 = ifa->ifa_next;

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

	/* 2. Unlink it */

	*ifap = ifa1->ifa_next;

	/* 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)
			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) {
		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;
	*ifap = ifa;

	/* 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) {
		in_dev = inetdev_init(dev);
		if (!in_dev) {
			inet_free_ifa(ifa);
			return -ENOBUFS;
		}
	}
	if (ifa->ifa_dev != in_dev) {
		BUG_TRAP(!ifa->ifa_dev);
		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;
}

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)
		goto out;
	__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;
	}
out:
	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;
	int rc = -EINVAL;

	ASSERT_RTNL();

	if (ifm->ifa_prefixlen > 32 || !rta[IFA_LOCAL - 1])
		goto out;

	rc = -ENODEV;
	if ((dev = __dev_get_by_index(ifm->ifa_index)) == NULL)
		goto out;

	rc = -ENOBUFS;
	if ((in_dev = __in_dev_get(dev)) == NULL) {
		in_dev = inetdev_init(dev);
		if (!in_dev)
			goto out;
	}

	if ((ifa = inet_alloc_ifa()) == NULL)
		goto out;

	if (!rta[IFA_ADDRESS - 1])
		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);

	rc = inet_insert_ifa(ifa);
out:
	return rc;
}

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

static __inline__ int inet_abc_len(u32 addr)
{
	int rc = -1;	/* Something else, probably a multicast. */

  	if (ZERONET(addr))
  		rc = 0;
	else {
		addr = ntohl(addr);

		if (IN_CLASSA(addr))
			rc = 8;
		else if (IN_CLASSB(addr))
			rc = 16;
		else if (IN_CLASSC(addr))
			rc = 24;
	}

  	return rc;
}


int devinet_ioctl(unsigned int cmd, void __user *arg)
{
	struct ifreq ifr;
	struct sockaddr_in sin_orig;
	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;