rtsock.c

早期freebsd实现

/*
 * Copyright (c) 1988, 1991, 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.
 *
 *	@(#)rtsock.c	8.3 (Berkeley) 1/4/94
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/domain.h>
#include <sys/protosw.h>

#include <net/if.h>
#include <net/route.h>
#include <net/raw_cb.h>

struct	sockaddr route_dst = { 2, PF_ROUTE, };
struct	sockaddr route_src = { 2, PF_ROUTE, };
struct	sockproto route_proto = { PF_ROUTE, };

struct walkarg {
	int	w_op, w_arg, w_given, w_needed, w_tmemsize;
	caddr_t	w_where, w_tmem;
};

static struct mbuf *
		rt_msg1 __P((int, struct rt_addrinfo *));
static int	rt_msg2 __P((int,
		    struct rt_addrinfo *, caddr_t, struct walkarg *));
static void	rt_xaddrs __P((caddr_t, caddr_t, struct rt_addrinfo *));

/* Sleazy use of local variables throughout file, warning!!!! */
#define dst	info.rti_info[RTAX_DST]
#define gate	info.rti_info[RTAX_GATEWAY]
#define netmask	info.rti_info[RTAX_NETMASK]
#define genmask	info.rti_info[RTAX_GENMASK]
#define ifpaddr	info.rti_info[RTAX_IFP]
#define ifaaddr	info.rti_info[RTAX_IFA]
#define brdaddr	info.rti_info[RTAX_BRD]

/*ARGSUSED*/
int
route_usrreq(so, req, m, nam, control)
	register struct socket *so;
	int req;
	struct mbuf *m, *nam, *control;
{
	register int error = 0;
	register struct rawcb *rp = sotorawcb(so);
	int s;

	if (req == PRU_ATTACH) {
		MALLOC(rp, struct rawcb *, sizeof(*rp), M_PCB, M_WAITOK);
		if (so->so_pcb = (caddr_t)rp)
			bzero(so->so_pcb, sizeof(*rp));

	}
	if (req == PRU_DETACH && rp) {
		int af = rp->rcb_proto.sp_protocol;
		if (af == AF_INET)
			route_cb.ip_count--;
		else if (af == AF_NS)
			route_cb.ns_count--;
		else if (af == AF_ISO)
			route_cb.iso_count--;
		route_cb.any_count--;
	}
	s = splnet();
	error = raw_usrreq(so, req, m, nam, control);
	rp = sotorawcb(so);
	if (req == PRU_ATTACH && rp) {
		int af = rp->rcb_proto.sp_protocol;
		if (error) {
			free((caddr_t)rp, M_PCB);
			splx(s);
			return (error);
		}
		if (af == AF_INET)
			route_cb.ip_count++;
		else if (af == AF_NS)
			route_cb.ns_count++;
		else if (af == AF_ISO)
			route_cb.iso_count++;
		rp->rcb_faddr = &route_src;
		route_cb.any_count++;
		soisconnected(so);
		so->so_options |= SO_USELOOPBACK;
	}
	splx(s);
	return (error);
}

/*ARGSUSED*/
int
route_output(m, so)
	register struct mbuf *m;
	struct socket *so;
{
	register struct rt_msghdr *rtm = 0;
	register struct rtentry *rt = 0;
	struct rtentry *saved_nrt = 0;
	struct rt_addrinfo info;
	int len, error = 0;
	struct ifnet *ifp = 0;
	struct ifaddr *ifa = 0;

#define senderr(e) { error = e; goto flush;}
	if (m == 0 || ((m->m_len < sizeof(long)) &&
		       (m = m_pullup(m, sizeof(long))) == 0))
		return (ENOBUFS);
	if ((m->m_flags & M_PKTHDR) == 0)
		panic("route_output");
	len = m->m_pkthdr.len;
	if (len < sizeof(*rtm) ||
	    len != mtod(m, struct rt_msghdr *)->rtm_msglen) {
		dst = 0;
		senderr(EINVAL);
	}
	R_Malloc(rtm, struct rt_msghdr *, len);
	if (rtm == 0) {
		dst = 0;
		senderr(ENOBUFS);
	}
	m_copydata(m, 0, len, (caddr_t)rtm);
	if (rtm->rtm_version != RTM_VERSION) {
		dst = 0;
		senderr(EPROTONOSUPPORT);
	}
	rtm->rtm_pid = curproc->p_pid;
	info.rti_addrs = rtm->rtm_addrs;
	rt_xaddrs((caddr_t)(rtm + 1), len + (caddr_t)rtm, &info);
	if (dst == 0)
		senderr(EINVAL);
	if (genmask) {
		struct radix_node *t;
		t = rn_addmask((caddr_t)genmask, 1, 2);
		if (t && Bcmp(genmask, t->rn_key, *(u_char *)genmask) == 0)
			genmask = (struct sockaddr *)(t->rn_key);
		else
			senderr(ENOBUFS);
	}
	switch (rtm->rtm_type) {

	case RTM_ADD:
		if (gate == 0)
			senderr(EINVAL);
		error = rtrequest(RTM_ADD, dst, gate, netmask,
					rtm->rtm_flags, &saved_nrt);
		if (error == 0 && saved_nrt) {
			rt_setmetrics(rtm->rtm_inits,
				&rtm->rtm_rmx, &saved_nrt->rt_rmx);
			saved_nrt->rt_refcnt--;
			saved_nrt->rt_genmask = genmask;
		}
		break;

	case RTM_DELETE:
		error = rtrequest(RTM_DELETE, dst, gate, netmask,
				rtm->rtm_flags, (struct rtentry **)0);
		break;

	case RTM_GET:
	case RTM_CHANGE:
	case RTM_LOCK:
		rt = rtalloc1(dst, 0);
		if (rt == 0)
			senderr(ESRCH);
		if (rtm->rtm_type != RTM_GET) {/* XXX: too grotty */
			struct radix_node *rn;
			extern struct radix_node_head *mask_rnhead;

			if (Bcmp(dst, rt_key(rt), dst->sa_len) != 0)
				senderr(ESRCH);
			if (netmask && (rn = rn_search(netmask,
					    mask_rnhead->rnh_treetop)))
				netmask = (struct sockaddr *)rn->rn_key;
			for (rn = rt->rt_nodes; rn; rn = rn->rn_dupedkey)
				if (netmask == (struct sockaddr *)rn->rn_mask)
					break;
			if (rn == 0)
				senderr(ETOOMANYREFS);
			rt = (struct rtentry *)rn;
		}
		switch(rtm->rtm_type) {

		case RTM_GET:
			dst = rt_key(rt);
			gate = rt->rt_gateway;
			netmask = rt_mask(rt);
			genmask = rt->rt_genmask;
			if (rtm->rtm_addrs & (RTA_IFP | RTA_IFA)) {
				if (ifp = rt->rt_ifp) {
					ifpaddr = ifp->if_addrlist->ifa_addr;
					ifaaddr = rt->rt_ifa->ifa_addr;
					rtm->rtm_index = ifp->if_index;
				} else {
					ifpaddr = 0;
					ifaaddr = 0;
			    }
			}
			len = rt_msg2(RTM_GET, &info, (caddr_t)0,
				(struct walkarg *)0);
			if (len > rtm->rtm_msglen) {
				struct rt_msghdr *new_rtm;
				R_Malloc(new_rtm, struct rt_msghdr *, len);
				if (new_rtm == 0)
					senderr(ENOBUFS);
				Bcopy(rtm, new_rtm, rtm->rtm_msglen);
				Free(rtm); rtm = new_rtm;
			}
			(void)rt_msg2(RTM_GET, &info, (caddr_t)rtm,
				(struct walkarg *)0);
			rtm->rtm_flags = rt->rt_flags;
			rtm->rtm_rmx = rt->rt_rmx;
			rtm->rtm_addrs = info.rti_addrs;
			break;

		case RTM_CHANGE:
			if (gate && rt_setgate(rt, rt_key(rt), gate))
				senderr(EDQUOT);
			/* new gateway could require new ifaddr, ifp;
			   flags may also be different; ifp may be specified
			   by ll sockaddr when protocol address is ambiguous */
			if (ifpaddr && (ifa = ifa_ifwithnet(ifpaddr)) &&
			    (ifp = ifa->ifa_ifp))
				ifa = ifaof_ifpforaddr(ifaaddr ? ifaaddr : gate,
							ifp);
			else if ((ifaaddr && (ifa = ifa_ifwithaddr(ifaaddr))) ||
				 (ifa = ifa_ifwithroute(rt->rt_flags,
							rt_key(rt), gate)))
				ifp = ifa->ifa_ifp;
			if (ifa) {
				register struct ifaddr *oifa = rt->rt_ifa;
				if (oifa != ifa) {
				    if (oifa && oifa->ifa_rtrequest)
					oifa->ifa_rtrequest(RTM_DELETE,
								rt, gate);
				    IFAFREE(rt->rt_ifa);
				    rt->rt_ifa = ifa;
				    ifa->ifa_refcnt++;
				    rt->rt_ifp = ifp;
				}
			}
			rt_setmetrics(rtm->rtm_inits, &rtm->rtm_rmx,
					&rt->rt_rmx);
			if (rt->rt_ifa && rt->rt_ifa->ifa_rtrequest)
			       rt->rt_ifa->ifa_rtrequest(RTM_ADD, rt, gate);
			if (genmask)
				rt->rt_genmask = genmask;
			/*
			 * Fall into
			 */
		case RTM_LOCK:
			rt->rt_rmx.rmx_locks &= ~(rtm->rtm_inits);
			rt->rt_rmx.rmx_locks |=
				(rtm->rtm_inits & rtm->rtm_rmx.rmx_locks);
			break;
		}
		break;

	default:
		senderr(EOPNOTSUPP);
	}

flush:
	if (rtm) {
		if (error)
			rtm->rtm_errno = error;
		else 
			rtm->rtm_flags |= RTF_DONE;
	}
	if (rt)
		rtfree(rt);
    {
	register struct rawcb *rp = 0;
	/*
	 * Check to see if we don't want our own messages.
	 */
	if ((so->so_options & SO_USELOOPBACK) == 0) {
		if (route_cb.any_count <= 1) {
			if (rtm)
				Free(rtm);
			m_freem(m);
			return (error);
		}
		/* There is another listener, so construct message */
		rp = sotorawcb(so);
	}
	if (rtm) {
		m_copyback(m, 0, rtm->rtm_msglen, (caddr_t)rtm);
		Free(rtm);
	}
	if (rp)
		rp->rcb_proto.sp_family = 0; /* Avoid us */
	if (dst)
		route_proto.sp_protocol = dst->sa_family;
	raw_input(m, &route_proto, &route_src, &route_dst);
	if (rp)
		rp->rcb_proto.sp_family = PF_ROUTE;
    }
	return (error);
}

void
rt_setmetrics(which, in, out)
	u_long which;
	register struct rt_metrics *in, *out;
{
#define metric(f, e) if (which & (f)) out->e = in->e;
	metric(RTV_RPIPE, rmx_recvpipe);
	metric(RTV_SPIPE, rmx_sendpipe);
	metric(RTV_SSTHRESH, rmx_ssthresh);
	metric(RTV_RTT, rmx_rtt);
	metric(RTV_RTTVAR, rmx_rttvar);
	metric(RTV_HOPCOUNT, rmx_hopcount);
	metric(RTV_MTU, rmx_mtu);
	metric(RTV_EXPIRE, rmx_expire);
#undef metric
}

#define ROUNDUP(a) \
	((a) > 0 ? (1 + (((a) - 1) | (sizeof(long) - 1))) : sizeof(long))
#define ADVANCE(x, n) (x += ROUNDUP((n)->sa_len))

static void
rt_xaddrs(cp, cplim, rtinfo)
	register caddr_t cp, cplim;
	register struct rt_addrinfo *rtinfo;
{
	register struct sockaddr *sa;
	register int i;

	bzero(rtinfo->rti_info, sizeof(rtinfo->rti_info));
	for (i = 0; (i < RTAX_MAX) && (cp < cplim); i++) {
		if ((rtinfo->rti_addrs & (1 << i)) == 0)
			continue;
		rtinfo->rti_info[i] = sa = (struct sockaddr *)cp;
		ADVANCE(cp, sa);
	}
}

/*
 * Copy data from a buffer back into the indicated mbuf chain,
 * starting "off" bytes from the beginning, extending the mbuf
 * chain if necessary.
 */
void
m_copyback(m0, off, len, cp)
	struct	mbuf *m0;
	register int off;
	register int len;
	caddr_t cp;
{
	register int mlen;
	register struct mbuf *m = m0, *n;
	int totlen = 0;

	if (m0 == 0)
		return;
	while (off > (mlen = m->m_len)) {
		off -= mlen;
		totlen += mlen;
		if (m->m_next == 0) {
			n = m_getclr(M_DONTWAIT, m->m_type);
			if (n == 0)
				goto out;
			n->m_len = min(MLEN, len + off);
			m->m_next = n;
		}
		m = m->m_next;
	}
	while (len > 0) {
		mlen = min (m->m_len - off, len);
		bcopy(cp, off + mtod(m, caddr_t), (unsigned)mlen);
		cp += mlen;
		len -= mlen;
		mlen += off;
		off = 0;
		totlen += mlen;
		if (len == 0)
			break;
		if (m->m_next == 0) {
			n = m_get(M_DONTWAIT, m->m_type);
			if (n == 0)
				break;