if.c

来自「基于组件方式开发操作系统的OSKIT源代码」· C语言 代码 · 共 1,096 行 · 第 1/2 页

/*
 * Copyright (c) 1980, 1986, 1993
 *	The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)if.c	8.3 (Berkeley) 1/4/94
 *	$Id: if.c,v 1.64 1998/12/16 18:30:42 phk Exp $
 */

#include "opt_compat.h"

#include <sys/param.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/protosw.h>
#include <sys/kernel.h>
#include <sys/sockio.h>
#include <sys/syslog.h>
#include <sys/sysctl.h>

#include <net/if.h>
#include <net/if_dl.h>
#include <net/radix.h>

/*
 * System initialization
 */

static int ifconf __P((u_long, caddr_t));
#ifdef OSKIT
void ifinit __P((void *));
#else
static void ifinit __P((void *));
#endif
static void if_qflush __P((struct ifqueue *));
static void if_slowtimo __P((void *));
static void link_rtrequest __P((int, struct rtentry *, struct sockaddr *));

#ifndef OSKIT
SYSINIT(interfaces, SI_SUB_PROTO_IF, SI_ORDER_FIRST, ifinit, NULL)
#endif /* !OSKIT */

MALLOC_DEFINE(M_IFADDR, "ifaddr", "interface address");
MALLOC_DEFINE(M_IFMADDR, "ether_multi", "link-level multicast address");

int	ifqmaxlen = IFQ_MAXLEN;
struct	ifnethead ifnet;	/* depend on static init XXX */

/*
 * Network interface utility routines.
 *
 * Routines with ifa_ifwith* names take sockaddr *'s as
 * parameters.
 *
 * This routine assumes that it will be called at splimp() or higher.
 */
/* ARGSUSED*/
void
ifinit(dummy)
	void *dummy;
{
	register struct ifnet *ifp;

	for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_link.tqe_next)
		if (ifp->if_snd.ifq_maxlen == 0)
			ifp->if_snd.ifq_maxlen = ifqmaxlen;
	if_slowtimo(0);
}

int if_index = 0;
struct ifaddr **ifnet_addrs;


/*
 * Attach an interface to the
 * list of "active" interfaces.
 */
void
if_attach(ifp)
	struct ifnet *ifp;
{
	unsigned socksize, ifasize;
	int namelen, masklen;
	char workbuf[64];
	register struct sockaddr_dl *sdl;
	register struct ifaddr *ifa;
	static int if_indexlim = 8;
	static int inited;

	if (!inited) {
		TAILQ_INIT(&ifnet);
		inited = 1;
	}

	TAILQ_INSERT_TAIL(&ifnet, ifp, if_link);
	ifp->if_index = ++if_index;
	/*
	 * XXX -
	 * The old code would work if the interface passed a pre-existing
	 * chain of ifaddrs to this code.  We don't trust our callers to
	 * properly initialize the tailq, however, so we no longer allow
	 * this unlikely case.
	 */
	TAILQ_INIT(&ifp->if_addrhead);
	LIST_INIT(&ifp->if_multiaddrs);
	getmicrotime(&ifp->if_lastchange);
	if (ifnet_addrs == 0 || if_index >= if_indexlim) {
		unsigned n = (if_indexlim <<= 1) * sizeof(ifa);
		struct ifaddr **q = (struct ifaddr **)
					malloc(n, M_IFADDR, M_WAITOK);
		bzero((caddr_t)q, n);
		if (ifnet_addrs) {
			bcopy((caddr_t)ifnet_addrs, (caddr_t)q, n/2);
			free((caddr_t)ifnet_addrs, M_IFADDR);
		}
		ifnet_addrs = q;
	}
	/*
	 * create a Link Level name for this device
	 */
	namelen = snprintf(workbuf, sizeof(workbuf),
	    "%s%d", ifp->if_name, ifp->if_unit);
#define _offsetof(t, m) ((int)((caddr_t)&((t *)0)->m))
	masklen = _offsetof(struct sockaddr_dl, sdl_data[0]) + namelen;
	socksize = masklen + ifp->if_addrlen;
#define ROUNDUP(a) (1 + (((a) - 1) | (sizeof(long) - 1)))
	if (socksize < sizeof(*sdl))
		socksize = sizeof(*sdl);
	socksize = ROUNDUP(socksize);
	ifasize = sizeof(*ifa) + 2 * socksize;
	ifa = (struct ifaddr *)malloc(ifasize, M_IFADDR, M_WAITOK);
	if (ifa) {
		bzero((caddr_t)ifa, ifasize);
		sdl = (struct sockaddr_dl *)(ifa + 1);
		sdl->sdl_len = socksize;
		sdl->sdl_family = AF_LINK;
		bcopy(workbuf, sdl->sdl_data, namelen);
		sdl->sdl_nlen = namelen;
		sdl->sdl_index = ifp->if_index;
		sdl->sdl_type = ifp->if_type;
		ifnet_addrs[if_index - 1] = ifa;
		ifa->ifa_ifp = ifp;
		ifa->ifa_rtrequest = link_rtrequest;
		ifa->ifa_addr = (struct sockaddr *)sdl;
		sdl = (struct sockaddr_dl *)(socksize + (caddr_t)sdl);
		ifa->ifa_netmask = (struct sockaddr *)sdl;
		sdl->sdl_len = masklen;
		while (namelen != 0)
			sdl->sdl_data[--namelen] = 0xff;
		TAILQ_INSERT_HEAD(&ifp->if_addrhead, ifa, ifa_link);
	}
}
/*
 * Locate an interface based on a complete address.
 */
/*ARGSUSED*/
struct ifaddr *
ifa_ifwithaddr(addr)
	register struct sockaddr *addr;
{
	register struct ifnet *ifp;
	register struct ifaddr *ifa;

#define	equal(a1, a2) \
  (bcmp((caddr_t)(a1), (caddr_t)(a2), ((struct sockaddr *)(a1))->sa_len) == 0)
	for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_link.tqe_next)
	    for (ifa = ifp->if_addrhead.tqh_first; ifa; 
		 ifa = ifa->ifa_link.tqe_next) {
		if (ifa->ifa_addr->sa_family != addr->sa_family)
			continue;
		if (equal(addr, ifa->ifa_addr))
			return (ifa);
		if ((ifp->if_flags & IFF_BROADCAST) && ifa->ifa_broadaddr &&
		    equal(ifa->ifa_broadaddr, addr))
			return (ifa);
	}
	return ((struct ifaddr *)0);
}
/*
 * Locate the point to point interface with a given destination address.
 */
/*ARGSUSED*/
struct ifaddr *
ifa_ifwithdstaddr(addr)
	register struct sockaddr *addr;
{
	register struct ifnet *ifp;
	register struct ifaddr *ifa;

	for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_link.tqe_next)
	    if (ifp->if_flags & IFF_POINTOPOINT)
		for (ifa = ifp->if_addrhead.tqh_first; ifa; 
		     ifa = ifa->ifa_link.tqe_next) {
			if (ifa->ifa_addr->sa_family != addr->sa_family)
				continue;
			if (ifa->ifa_dstaddr && equal(addr, ifa->ifa_dstaddr))
				return (ifa);
	}
	return ((struct ifaddr *)0);
}

/*
 * Find an interface on a specific network.  If many, choice
 * is most specific found.
 */
struct ifaddr *
ifa_ifwithnet(addr)
	struct sockaddr *addr;
{
	register struct ifnet *ifp;
	register struct ifaddr *ifa;
	struct ifaddr *ifa_maybe = (struct ifaddr *) 0;
	u_int af = addr->sa_family;
	char *addr_data = addr->sa_data, *cplim;

	/*
	 * AF_LINK addresses can be looked up directly by their index number,
	 * so do that if we can.
	 */
	if (af == AF_LINK) {
	    register struct sockaddr_dl *sdl = (struct sockaddr_dl *)addr;
	    if (sdl->sdl_index && sdl->sdl_index <= if_index)
		return (ifnet_addrs[sdl->sdl_index - 1]);
	}

	/* 
	 * Scan though each interface, looking for ones that have
	 * addresses in this address family.
	 */
	for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_link.tqe_next) {
		for (ifa = ifp->if_addrhead.tqh_first; ifa;
		     ifa = ifa->ifa_link.tqe_next) {
			register char *cp, *cp2, *cp3;

			if (ifa->ifa_addr->sa_family != af)
next:				continue;
			if (ifp->if_flags & IFF_POINTOPOINT) {
				/*
				 * This is a bit broken as it doesn't 
				 * take into account that the remote end may 
				 * be a single node in the network we are
				 * looking for.
				 * The trouble is that we don't know the 
				 * netmask for the remote end.
				 */
				if (ifa->ifa_dstaddr != 0
				    && equal(addr, ifa->ifa_dstaddr))
 					return (ifa);
			} else {
				/*
				 * if we have a special address handler,
				 * then use it instead of the generic one.
				 */
	          		if (ifa->ifa_claim_addr) {
					if ((*ifa->ifa_claim_addr)(ifa, addr)) {
						return (ifa);
					} else {
						continue;
					}
				}

				/*
				 * Scan all the bits in the ifa's address.
				 * If a bit dissagrees with what we are
				 * looking for, mask it with the netmask
				 * to see if it really matters.
				 * (A byte at a time)
				 */
				if (ifa->ifa_netmask == 0)
					continue;
				cp = addr_data;
				cp2 = ifa->ifa_addr->sa_data;
				cp3 = ifa->ifa_netmask->sa_data;
				cplim = ifa->ifa_netmask->sa_len
					+ (char *)ifa->ifa_netmask;
				while (cp3 < cplim)
					if ((*cp++ ^ *cp2++) & *cp3++)
						goto next; /* next address! */
				/*
				 * If the netmask of what we just found
				 * is more specific than what we had before
				 * (if we had one) then remember the new one
				 * before continuing to search
				 * for an even better one.
				 */
				if (ifa_maybe == 0 ||
				    rn_refines((caddr_t)ifa->ifa_netmask,
				    (caddr_t)ifa_maybe->ifa_netmask))
					ifa_maybe = ifa;
			}
		}
	}
	return (ifa_maybe);
}

/*
 * Find an interface address specific to an interface best matching
 * a given address.
 */
struct ifaddr *
ifaof_ifpforaddr(addr, ifp)
	struct sockaddr *addr;
	register struct ifnet *ifp;
{
	register struct ifaddr *ifa;
	register char *cp, *cp2, *cp3;
	register char *cplim;
	struct ifaddr *ifa_maybe = 0;
	u_int af = addr->sa_family;

	if (af >= AF_MAX)
		return (0);
	for (ifa = ifp->if_addrhead.tqh_first; ifa; 
	     ifa = ifa->ifa_link.tqe_next) {
		if (ifa->ifa_addr->sa_family != af)
			continue;
		if (ifa_maybe == 0)
			ifa_maybe = ifa;
		if (ifa->ifa_netmask == 0) {
			if (equal(addr, ifa->ifa_addr) ||
			    (ifa->ifa_dstaddr && equal(addr, ifa->ifa_dstaddr)))
				return (ifa);
			continue;
		}
		if (ifp->if_flags & IFF_POINTOPOINT) {
			if (equal(addr, ifa->ifa_dstaddr))
				return (ifa);
		} else {
			cp = addr->sa_data;
			cp2 = ifa->ifa_addr->sa_data;
			cp3 = ifa->ifa_netmask->sa_data;
			cplim = ifa->ifa_netmask->sa_len + (char *)ifa->ifa_netmask;
			for (; cp3 < cplim; cp3++)
				if ((*cp++ ^ *cp2++) & *cp3)
					break;
			if (cp3 == cplim)
				return (ifa);
		}
	}
	return (ifa_maybe);
}

#include <net/route.h>

/*
 * Default action when installing a route with a Link Level gateway.
 * Lookup an appropriate real ifa to point to.
 * This should be moved to /sys/net/link.c eventually.
 */
static void
link_rtrequest(cmd, rt, sa)
	int cmd;
	register struct rtentry *rt;
	struct sockaddr *sa;
{
	register struct ifaddr *ifa;
	struct sockaddr *dst;
	struct ifnet *ifp;

	if (cmd != RTM_ADD || ((ifa = rt->rt_ifa) == 0) ||
	    ((ifp = ifa->ifa_ifp) == 0) || ((dst = rt_key(rt)) == 0))
		return;
	ifa = ifaof_ifpforaddr(dst, ifp);
	if (ifa) {
		IFAFREE(rt->rt_ifa);
		rt->rt_ifa = ifa;
		ifa->ifa_refcnt++;
		if (ifa->ifa_rtrequest && ifa->ifa_rtrequest != link_rtrequest)
			ifa->ifa_rtrequest(cmd, rt, sa);
	}
}

/*
 * Mark an interface down and notify protocols of
 * the transition.
 * NOTE: must be called at splnet or eqivalent.
 */
void
if_unroute(ifp, flag, fam)
	register struct ifnet *ifp;
	int flag, fam;
{
	register struct ifaddr *ifa;

	ifp->if_flags &= ~flag;
	getmicrotime(&ifp->if_lastchange);
	TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
		if (fam == PF_UNSPEC || (fam == ifa->ifa_addr->sa_family))
			pfctlinput(PRC_IFDOWN, ifa->ifa_addr);
	if_qflush(&ifp->if_snd);
	rt_ifmsg(ifp);
}

/*
 * Mark an interface up and notify protocols of
 * the transition.
 * NOTE: must be called at splnet or eqivalent.
 */
void
if_route(ifp, flag, fam)
	register struct ifnet *ifp;
	int flag, fam;
{
	register struct ifaddr *ifa;

	ifp->if_flags |= flag;
	getmicrotime(&ifp->if_lastchange);
	TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
		if (fam == PF_UNSPEC || (fam == ifa->ifa_addr->sa_family))
			pfctlinput(PRC_IFUP, ifa->ifa_addr);
	rt_ifmsg(ifp);
}

/*
 * Mark an interface down and notify protocols of
 * the transition.
 * NOTE: must be called at splnet or eqivalent.
 */
void
if_down(ifp)
	register struct ifnet *ifp;
{

	if_unroute(ifp, IFF_UP, AF_UNSPEC);
}

/*
 * Mark an interface up and notify protocols of
 * the transition.
 * NOTE: must be called at splnet or eqivalent.
 */
void
if_up(ifp)
	register struct ifnet *ifp;
{

	if_route(ifp, IFF_UP, AF_UNSPEC);
}

/*
 * Flush an interface queue.
 */
static void
if_qflush(ifq)
	register struct ifqueue *ifq;
{
	register struct mbuf *m, *n;

	n = ifq->ifq_head;
	while ((m = n) != 0) {
		n = m->m_act;
		m_freem(m);
	}
	ifq->ifq_head = 0;
	ifq->ifq_tail = 0;
	ifq->ifq_len = 0;
}

/*
 * Handle interface watchdog timer routines.  Called
 * from softclock, we decrement timers (if set) and
 * call the appropriate interface routine on expiration.
 */
static void
if_slowtimo(arg)
	void *arg;
{
	register struct ifnet *ifp;
	int s = splimp();

	for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_link.tqe_next) {
		if (ifp->if_timer == 0 || --ifp->if_timer)
			continue;
		if (ifp->if_watchdog)
			(*ifp->if_watchdog)(ifp);
	}
	splx(s);
	timeout(if_slowtimo, (void *)0, hz / IFNET_SLOWHZ);
}

/*
 * Map interface name to
 * interface structure pointer.
 */
struct ifnet *
ifunit(name)
	register char *name;
{
	char namebuf[IFNAMSIZ + 1];
	register char *cp, *cp2;
	char *end;
	register struct ifnet *ifp;
	int unit;
	unsigned len;
	register char c = '\0';

	/*
	 * Look for a non numeric part
	 */
	end = name + IFNAMSIZ; 
	cp2 = namebuf;
	cp = name; 
	while ((cp < end) && (c = *cp)) {
		if (c >= '0' && c <= '9')
			break;
		*cp2++ = c;
		cp++;
	}
	if ((cp == end) || (c == '\0') || (cp == name))
		return ((struct ifnet *)0);
	*cp2 = '\0';
	/*
	 * check we have a legal number (limit to 7 digits?)
	 */
	len = cp - name + 1;
	for (unit = 0;
	    ((c = *cp) >= '0') && (c <= '9') && (unit < 1000000); cp++ )