arp.c

GNU Hurd 源代码

/* linux/net/inet/arp.c
 *
 * Version:	$Id: arp.c,v 1.77.2.4 1999/09/23 19:03:36 davem Exp $
 *
 * Copyright (C) 1994 by Florian  La Roche
 *
 * This module implements the Address Resolution Protocol ARP (RFC 826),
 * which is used to convert IP addresses (or in the future maybe other
 * high-level addresses) into a low-level hardware address (like an Ethernet
 * address).
 *
 * 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:
 *		Alan Cox	:	Removed the Ethernet assumptions in
 *					Florian's code
 *		Alan Cox	:	Fixed some small errors in the ARP
 *					logic
 *		Alan Cox	:	Allow >4K in /proc
 *		Alan Cox	:	Make ARP add its own protocol entry
 *		Ross Martin     :       Rewrote arp_rcv() and arp_get_info()
 *		Stephen Henson	:	Add AX25 support to arp_get_info()
 *		Alan Cox	:	Drop data when a device is downed.
 *		Alan Cox	:	Use init_timer().
 *		Alan Cox	:	Double lock fixes.
 *		Martin Seine	:	Move the arphdr structure
 *					to if_arp.h for compatibility.
 *					with BSD based programs.
 *		Andrew Tridgell :       Added ARP netmask code and
 *					re-arranged proxy handling.
 *		Alan Cox	:	Changed to use notifiers.
 *		Niibe Yutaka	:	Reply for this device or proxies only.
 *		Alan Cox	:	Don't proxy across hardware types!
 *		Jonathan Naylor :	Added support for NET/ROM.
 *		Mike Shaver     :       RFC1122 checks.
 *		Jonathan Naylor :	Only lookup the hardware address for
 *					the correct hardware type.
 *		Germano Caronni	:	Assorted subtle races.
 *		Craig Schlenter :	Don't modify permanent entry
 *					during arp_rcv.
 *		Russ Nelson	:	Tidied up a few bits.
 *		Alexey Kuznetsov:	Major changes to caching and behaviour,
 *					eg intelligent arp probing and
 *					generation
 *					of host down events.
 *		Alan Cox	:	Missing unlock in device events.
 *		Eckes		:	ARP ioctl control errors.
 *		Alexey Kuznetsov:	Arp free fix.
 *		Manuel Rodriguez:	Gratuitous ARP.
 *              Jonathan Layes  :       Added arpd support through kerneld
 *                                      message queue (960314)
 *		Mike Shaver	:	/proc/sys/net/ipv4/arp_* support
 *		Mike McLagan    :	Routing by source
 *		Stuart Cheshire	:	Metricom and grat arp fixes
 *					*** FOR 2.1 clean this up ***
 *		Lawrence V. Stefani: (08/12/96) Added FDDI support.
 *		Alan Cox 	:	Took the AP1000 nasty FDDI hack and
 *					folded into the mainstream FDDI code.
 *					Ack spit, Linus how did you allow that
 *					one in...
 *		Jes Sorensen	:	Make FDDI work again in 2.1.x and
 *					clean up the APFDDI & gen. FDDI bits.
 *		Alexey Kuznetsov:	new arp state machine;
 *					now it is in net/core/neighbour.c.
 *		Julian Anastasov:	"hidden" flag: hide the
 *					interface and don't reply for it
 */

/* RFC1122 Status:
   2.3.2.1 (ARP Cache Validation):
     MUST provide mechanism to flush stale cache entries (OK)
     SHOULD be able to configure cache timeout (OK)
     MUST throttle ARP retransmits (OK)
   2.3.2.2 (ARP Packet Queue):
     SHOULD save at least one packet from each "conversation" with an
       unresolved IP address.  (OK)
   950727 -- MS
*/

#include <linux/types.h>
#include <linux/string.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/config.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/errno.h>
#include <linux/in.h>
#include <linux/mm.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/fddidevice.h>
#include <linux/if_arp.h>
#include <linux/trdevice.h>
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <linux/stat.h>
#include <linux/init.h>
#ifdef CONFIG_SYSCTL
#include <linux/sysctl.h>
#endif

#include <net/ip.h>
#include <net/icmp.h>
#include <net/route.h>
#include <net/protocol.h>
#include <net/tcp.h>
#include <net/sock.h>
#include <net/arp.h>
#if defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE)
#include <net/ax25.h>
#if defined(CONFIG_NETROM) || defined(CONFIG_NETROM_MODULE)
#include <net/netrom.h>
#endif
#endif

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

/*
 *	Interface to generic neighbour cache.
 */
static int arp_constructor(struct neighbour *neigh);
static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb);
static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb);
static void parp_redo(struct sk_buff *skb);

static struct neigh_ops arp_generic_ops =
{
	AF_INET,
	NULL,
	arp_solicit,
	arp_error_report,
	neigh_resolve_output,
	neigh_connected_output,
	dev_queue_xmit,
	dev_queue_xmit
};

static struct neigh_ops arp_hh_ops =
{
	AF_INET,
	NULL,
	arp_solicit,
	arp_error_report,
	neigh_resolve_output,
	neigh_resolve_output,
	dev_queue_xmit,
	dev_queue_xmit
};

static struct neigh_ops arp_direct_ops =
{
	AF_INET,
	NULL,
	NULL,
	NULL,
	dev_queue_xmit,
	dev_queue_xmit,
	dev_queue_xmit,
	dev_queue_xmit
};

struct neigh_ops arp_broken_ops =
{
	AF_INET,
	NULL,
	arp_solicit,
	arp_error_report,
	neigh_compat_output,
	neigh_compat_output,
	dev_queue_xmit,
	dev_queue_xmit,
};

struct neigh_table arp_tbl =
{
	NULL,
	AF_INET,
	sizeof(struct neighbour) + 4,
	4,
	arp_constructor,
	NULL,
	NULL,
	parp_redo,
        { NULL, NULL, &arp_tbl, 0, NULL, NULL,
		  30*HZ, 1*HZ, 60*HZ, 30*HZ, 5*HZ, 3, 3, 0, 3, 1*HZ, (8*HZ)/10, 64, 1*HZ },
	30*HZ, 128, 512, 1024,
};

int arp_mc_map(u32 addr, u8 *haddr, struct device *dev, int dir)
{
	switch (dev->type) {
	case ARPHRD_ETHER:
	case ARPHRD_FDDI:
		ip_eth_mc_map(addr, haddr) ; 
		return 0 ; 
	case ARPHRD_IEEE802:
		if ( (dev->name[0] == 't') && (dev->name[1] == 'r')) /* Token Ring */
			ip_tr_mc_map(addr,haddr) ; 
		else  
			ip_eth_mc_map(addr, haddr);
		return 0;
	default:
		if (dir) {
			memcpy(haddr, dev->broadcast, dev->addr_len);
			return 0;
		}
	}
	return -EINVAL;
}



static int arp_constructor(struct neighbour *neigh)
{
	u32 addr = *(u32*)neigh->primary_key;
	struct device *dev = neigh->dev;
	struct in_device *in_dev = dev->ip_ptr;

	if (in_dev == NULL)
		return -EINVAL;

	neigh->type = inet_addr_type(addr);
	if (in_dev->arp_parms)
		neigh->parms = in_dev->arp_parms;

	if (dev->hard_header == NULL) {
		neigh->nud_state = NUD_NOARP;
		neigh->ops = &arp_direct_ops;
		neigh->output = neigh->ops->queue_xmit;
	} else {
		/* Good devices (checked by reading texts, but only Ethernet is
		   tested)

		   ARPHRD_ETHER: (ethernet, apfddi)
		   ARPHRD_FDDI: (fddi)
		   ARPHRD_IEEE802: (tr)
		   ARPHRD_METRICOM: (strip)
		   ARPHRD_ARCNET:
		   etc. etc. etc.

		   ARPHRD_IPDDP will also work, if author repairs it.
		   I did not it, because this driver does not work even
		   in old paradigm.
		 */

#if 1
		/* So... these "amateur" devices are hopeless.
		   The only thing, that I can say now:
		   It is very sad that we need to keep ugly obsolete
		   code to make them happy.

		   They should be moved to more reasonable state, now
		   they use rebuild_header INSTEAD OF hard_start_xmit!!!
		   Besides that, they are sort of out of date
		   (a lot of redundant clones/copies, useless in 2.1),
		   I wonder why people believe that they work.
		 */
		switch (dev->type) {
		default:
			break;
		case ARPHRD_ROSE:
#if defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE)
		case ARPHRD_AX25:
#if defined(CONFIG_NETROM) || defined(CONFIG_NETROM_MODULE)
		case ARPHRD_NETROM:
#endif
			neigh->ops = &arp_broken_ops;
			neigh->output = neigh->ops->output;
			return 0;
#endif
			break;
		}
#endif
		if (neigh->type == RTN_MULTICAST) {
			neigh->nud_state = NUD_NOARP;
			arp_mc_map(addr, neigh->ha, dev, 1);
		} else if (dev->flags&(IFF_NOARP|IFF_LOOPBACK)) {
			neigh->nud_state = NUD_NOARP;
			memcpy(neigh->ha, dev->dev_addr, dev->addr_len);
		} else if (neigh->type == RTN_BROADCAST || dev->flags&IFF_POINTOPOINT) {
			neigh->nud_state = NUD_NOARP;
			memcpy(neigh->ha, dev->broadcast, dev->addr_len);
		}
		if (dev->hard_header_cache)
			neigh->ops = &arp_hh_ops;
		else
			neigh->ops = &arp_generic_ops;
		if (neigh->nud_state&NUD_VALID)
			neigh->output = neigh->ops->connected_output;
		else
			neigh->output = neigh->ops->output;
	}

	return 0;
}

static void arp_error_report(struct neighbour *neigh, struct sk_buff *skb)
{
	dst_link_failure(skb);
	kfree_skb(skb);
}

static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb)
{
	u32 saddr;
	u8  *dst_ha = NULL;
	struct device *dev = neigh->dev;
	struct device *dev2;
	struct in_device *in_dev2;
	u32 target = *(u32*)neigh->primary_key;
	int probes = neigh->probes;

	if (skb &&
	    (dev2 = ip_dev_find(skb->nh.iph->saddr)) != NULL &&
	    (in_dev2 = dev2->ip_ptr) != NULL &&
	    !IN_DEV_HIDDEN(in_dev2))
		saddr = skb->nh.iph->saddr;
	else
		saddr = inet_select_addr(dev, target, RT_SCOPE_LINK);

	if ((probes -= neigh->parms->ucast_probes) < 0) {
		if (!(neigh->nud_state&NUD_VALID))
			printk(KERN_DEBUG "trying to ucast probe in NUD_INVALID\n");
		dst_ha = neigh->ha;
	} else if ((probes -= neigh->parms->app_probes) < 0) {
#ifdef CONFIG_ARPD
		neigh_app_ns(neigh);
#endif
		return;
	}

	arp_send(ARPOP_REQUEST, ETH_P_ARP, target, dev, saddr,
		 dst_ha, dev->dev_addr, NULL);
}

/* OBSOLETE FUNCTIONS */

/*
 *	Find an arp mapping in the cache. If not found, post a request.
 *
 *	It is very UGLY routine: it DOES NOT use skb->dst->neighbour,
 *	even if it exists. It is supposed that skb->dev was mangled
 *	by a virtual device (eql, shaper). Nobody but broken devices
 *	is allowed to use this function, it is scheduled to be removed. --ANK
 */

static int arp_set_predefined(int addr_hint, unsigned char * haddr, u32 paddr, struct device * dev)
{
	switch (addr_hint) {
	case RTN_LOCAL:
		printk(KERN_DEBUG "ARP: arp called for own IP address\n");
		memcpy(haddr, dev->dev_addr, dev->addr_len);
		return 1;
	case RTN_MULTICAST:
		arp_mc_map(paddr, haddr, dev, 1);
		return 1;
	case RTN_BROADCAST:
		memcpy(haddr, dev->broadcast, dev->addr_len);
		return 1;
	}
	return 0;
}


int arp_find(unsigned char *haddr, struct sk_buff *skb)
{
	struct device *dev = skb->dev;
	u32 paddr;
	struct neighbour *n;

	if (!skb->dst) {
		printk(KERN_DEBUG "arp_find is called with dst==NULL\n");
		kfree_skb(skb);
		return 1;
	}

	paddr = ((struct rtable*)skb->dst)->rt_gateway;

	if (arp_set_predefined(inet_addr_type(paddr), haddr, paddr, dev))
		return 0;

	start_bh_atomic();
	n = __neigh_lookup(&arp_tbl, &paddr, dev, 1);

	if (n) {
		n->used = jiffies;
		if (n->nud_state&NUD_VALID || neigh_event_send(n, skb) == 0) {
			memcpy(haddr, n->ha, dev->addr_len);
			neigh_release(n);
			end_bh_atomic();
			return 0;
		}
		neigh_release(n);
	} else
		kfree_skb(skb);
	end_bh_atomic();
	return 1;
}

/* END OF OBSOLETE FUNCTIONS */

/*
 * Note: requires bh_atomic locking.
 */
int arp_bind_neighbour(struct dst_entry *dst)
{
	struct device *dev = dst->dev;

	if (dev == NULL)
		return 0;
	if (dst->neighbour == NULL) {
		u32 nexthop = ((struct rtable*)dst)->rt_gateway;
		if (dev->flags&(IFF_LOOPBACK|IFF_POINTOPOINT))
			nexthop = 0;
		dst->neighbour = __neigh_lookup(&arp_tbl, &nexthop, dev, 1);
	}
	return (dst->neighbour != NULL);
}

/*
 *	Interface to link layer: send routine and receive handler.
 */

/*
 *	Create and send an arp packet. If (dest_hw == NULL), we create a broadcast
 *	message.
 */

void arp_send(int type, int ptype, u32 dest_ip,
	      struct device *dev, u32 src_ip,
	      unsigned char *dest_hw, unsigned char *src_hw,
	      unsigned char *target_hw)
{
	struct sk_buff *skb;
	struct arphdr *arp;
	unsigned char *arp_ptr;

	/*
	 *	No arp on this interface.
	 */

	if (dev->flags&IFF_NOARP)
		return;

	/*
	 *	Allocate a buffer
	 */

	skb = alloc_skb(sizeof(struct arphdr)+ 2*(dev->addr_len+4)
				+ dev->hard_header_len + 15, GFP_ATOMIC);
	if (skb == NULL)
		return;

	skb_reserve(skb, (dev->hard_header_len+15)&~15);
	skb->nh.raw = skb->data;
	arp = (struct arphdr *) skb_put(skb,sizeof(struct arphdr) + 2*(dev->addr_len+4));
	skb->dev = dev;
	skb->protocol = __constant_htons (ETH_P_ARP);
	if (src_hw == NULL)
		src_hw = dev->dev_addr;
	if (dest_hw == NULL)
		dest_hw = dev->broadcast;

	/*
	 *	Fill the device header for the ARP frame
	 */
	dev->hard_header(skb,dev,ptype,dest_hw,src_hw,skb->len);

	/*
	 * Fill out the arp protocol part.
	 *
	 * The arp hardware type should match the device type, except for FDDI,
	 * which (according to RFC 1390) should always equal 1 (Ethernet).
	 */
	/*
	 *	Exceptions everywhere. AX.25 uses the AX.25 PID value not the
	 *	DIX code for the protocol. Make these device structure fields.
	 */
	switch (dev->type) {
	default:
		arp->ar_hrd = htons(dev->type);
		arp->ar_pro = __constant_htons(ETH_P_IP);
		break;

#if defined(CONFIG_AX25) || defined(CONFIG_AX25_MODULE)
	case ARPHRD_AX25:
		arp->ar_hrd = __constant_htons(ARPHRD_AX25);
		arp->ar_pro = __constant_htons(AX25_P_IP);
		break;

#if defined(CONFIG_NETROM) || defined(CONFIG_NETROM_MODULE)
	case ARPHRD_NETROM:
		arp->ar_hrd = __constant_htons(ARPHRD_NETROM);
		arp->ar_pro = __constant_htons(AX25_P_IP);
		break;
#endif
#endif

#ifdef CONFIG_FDDI
	case ARPHRD_FDDI:
		arp->ar_hrd = __constant_htons(ARPHRD_ETHER);
		arp->ar_pro = __constant_htons(ETH_P_IP);
		break;
#endif
	}

	arp->ar_hln = dev->addr_len;
	arp->ar_pln = 4;
	arp->ar_op = htons(type);

	arp_ptr=(unsigned char *)(arp+1);

	memcpy(arp_ptr, src_hw, dev->addr_len);
	arp_ptr+=dev->addr_len;
	memcpy(arp_ptr, &src_ip,4);
	arp_ptr+=4;
	if (target_hw != NULL)
		memcpy(arp_ptr, target_hw, dev->addr_len);
	else
		memset(arp_ptr, 0, dev->addr_len);
	arp_ptr+=dev->addr_len;
	memcpy(arp_ptr, &dest_ip, 4);
	skb->dev = dev;

	dev_queue_xmit(skb);
}

static void parp_redo(struct sk_buff *skb)
{
	arp_rcv(skb, skb->dev, NULL);
}

/*
 *	Receive an arp request by the device layer.
 */

int arp_rcv(struct sk_buff *skb, struct device *dev, struct packet_type *pt)
{
	struct arphdr *arp = skb->nh.arph;
	unsigned char *arp_ptr= (unsigned char *)(arp+1);
	struct rtable *rt;
	unsigned char *sha, *tha;
	u32 sip, tip;
	u16 dev_type = dev->type;
	int addr_type;
	struct in_device *in_dev = dev->ip_ptr;
	struct neighbour *n;

/*
 *	The hardware length of the packet should match the hardware length
 *	of the device.  Similarly, the hardware types should match.  The
 *	device should be ARP-able.  Also, if pln is not 4, then the lookup
 *	is not from an IP number.  We can't currently handle this, so toss
 *	it.
 */
	if (in_dev == NULL ||
	    arp->ar_hln != dev->addr_len    ||
	    dev->flags & IFF_NOARP ||
	    skb->pkt_type == PACKET_OTHERHOST ||
	    skb->pkt_type == PACKET_LOOPBACK ||
	    arp->ar_pln != 4)
		goto out;

	switch (dev_type) {
	default:
		if (arp->ar_pro != __constant_htons(ETH_P_IP))
			goto out;
		if (htons(dev_type) != arp->ar_hrd)
			goto out;
		break;
#ifdef CONFIG_NET_ETHERNET
	case ARPHRD_ETHER:
		/*
		 * ETHERNET devices will accept ARP hardware types of either
		 * 1 (Ethernet) or 6 (IEEE 802.2).
		 */
		if (arp->ar_hrd != __constant_htons(ARPHRD_ETHER) &&
		    arp->ar_hrd != __constant_htons(ARPHRD_IEEE802))
			goto out;
		if (arp->ar_pro != __constant_htons(ETH_P_IP))
			goto out;
		break;
#endif
#ifdef CONFIG_FDDI
	case ARPHRD_FDDI:
		/*
		 * According to RFC 1390, FDDI devices should accept ARP hardware types
		 * of 1 (Ethernet).  However, to be more robust, we'll accept hardware
		 * types of either 1 (Ethernet) or 6 (IEEE 802.2).
		 */