freebsd-skq/sys/netipx/ipx_usrreq.c

617 lines
14 KiB
C
Raw Normal View History

/*
* Copyright (c) 1995, Mike Mitchell
* Copyright (c) 1984, 1985, 1986, 1987, 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.
*
* @(#)ipx_usrreq.c
*/
2003-06-11 05:37:42 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_ipx.h"
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/mbuf.h>
#include <sys/protosw.h>
#include <sys/signalvar.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/sx.h>
#include <sys/sysctl.h>
#include <sys/systm.h>
#include <net/if.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netipx/ipx.h>
#include <netipx/ipx_if.h>
#include <netipx/ipx_ip.h>
#include <netipx/ipx_pcb.h>
#include <netipx/ipx_var.h>
/*
* IPX protocol implementation.
*/
1998-02-09 06:11:36 +00:00
static int ipxsendspace = IPXSNDQ;
SYSCTL_INT(_net_ipx_ipx, OID_AUTO, ipxsendspace, CTLFLAG_RW,
&ipxsendspace, 0, "");
1998-02-09 06:11:36 +00:00
static int ipxrecvspace = IPXRCVQ;
SYSCTL_INT(_net_ipx_ipx, OID_AUTO, ipxrecvspace, CTLFLAG_RW,
&ipxrecvspace, 0, "");
static int ipx_usr_abort(struct socket *so);
static int ipx_attach(struct socket *so, int proto, struct thread *td);
static int ipx_bind(struct socket *so, struct sockaddr *nam, struct thread *td);
static int ipx_connect(struct socket *so, struct sockaddr *nam,
struct thread *td);
static int ipx_detach(struct socket *so);
static int ipx_disconnect(struct socket *so);
static int ipx_send(struct socket *so, int flags, struct mbuf *m,
struct sockaddr *addr, struct mbuf *control,
struct thread *td);
static int ipx_shutdown(struct socket *so);
static int ripx_attach(struct socket *so, int proto, struct thread *td);
static int ipx_output(struct ipxpcb *ipxp, struct mbuf *m0);
struct pr_usrreqs ipx_usrreqs = {
ipx_usr_abort, pru_accept_notsupp, ipx_attach, ipx_bind,
ipx_connect, pru_connect2_notsupp, ipx_control, ipx_detach,
ipx_disconnect, pru_listen_notsupp, ipx_peeraddr, pru_rcvd_notsupp,
pru_rcvoob_notsupp, ipx_send, pru_sense_null, ipx_shutdown,
ipx_sockaddr, sosend, soreceive, sopoll
};
struct pr_usrreqs ripx_usrreqs = {
ipx_usr_abort, pru_accept_notsupp, ripx_attach, ipx_bind,
ipx_connect, pru_connect2_notsupp, ipx_control, ipx_detach,
ipx_disconnect, pru_listen_notsupp, ipx_peeraddr, pru_rcvd_notsupp,
pru_rcvoob_notsupp, ipx_send, pru_sense_null, ipx_shutdown,
ipx_sockaddr, sosend, soreceive, sopoll
};
/*
* This may also be called for raw listeners.
*/
void
ipx_input(m, ipxp)
struct mbuf *m;
register struct ipxpcb *ipxp;
{
register struct ipx *ipx = mtod(m, struct ipx *);
struct ifnet *ifp = m->m_pkthdr.rcvif;
struct sockaddr_ipx ipx_ipx;
if (ipxp == NULL)
panic("No ipxpcb");
/*
* Construct sockaddr format source address.
* Stuff source address and datagram in user buffer.
*/
ipx_ipx.sipx_len = sizeof(ipx_ipx);
ipx_ipx.sipx_family = AF_IPX;
ipx_ipx.sipx_addr = ipx->ipx_sna;
ipx_ipx.sipx_zero[0] = '\0';
ipx_ipx.sipx_zero[1] = '\0';
if (ipx_neteqnn(ipx->ipx_sna.x_net, ipx_zeronet) && ifp != NULL) {
register struct ifaddr *ifa;
for (ifa = TAILQ_FIRST(&ifp->if_addrhead); ifa != NULL;
ifa = TAILQ_NEXT(ifa, ifa_link)) {
if (ifa->ifa_addr->sa_family == AF_IPX) {
ipx_ipx.sipx_addr.x_net =
IA_SIPX(ifa)->sipx_addr.x_net;
break;
}
}
}
ipxp->ipxp_rpt = ipx->ipx_pt;
if (!(ipxp->ipxp_flags & IPXP_RAWIN) ) {
m->m_len -= sizeof(struct ipx);
m->m_pkthdr.len -= sizeof(struct ipx);
m->m_data += sizeof(struct ipx);
}
if (sbappendaddr(&ipxp->ipxp_socket->so_rcv, (struct sockaddr *)&ipx_ipx,
m, (struct mbuf *)NULL) == 0)
goto bad;
sorwakeup(ipxp->ipxp_socket);
return;
bad:
m_freem(m);
}
void
ipx_abort(ipxp)
struct ipxpcb *ipxp;
{
struct socket *so = ipxp->ipxp_socket;
ipx_pcbdisconnect(ipxp);
soisdisconnected(so);
}
/*
* Drop connection, reporting
* the specified error.
*/
void
ipx_drop(ipxp, errno)
register struct ipxpcb *ipxp;
int errno;
{
struct socket *so = ipxp->ipxp_socket;
/*
* someday, in the IPX world
* we will generate error protocol packets
* announcing that the socket has gone away.
*
* XXX Probably never. IPX does not have error packets.
*/
/*if (TCPS_HAVERCVDSYN(tp->t_state)) {
tp->t_state = TCPS_CLOSED;
tcp_output(tp);
}*/
so->so_error = errno;
ipx_pcbdisconnect(ipxp);
soisdisconnected(so);
}
static int
ipx_output(ipxp, m0)
struct ipxpcb *ipxp;
struct mbuf *m0;
{
register struct ipx *ipx;
register struct socket *so;
register int len = 0;
register struct route *ro;
struct mbuf *m;
struct mbuf *mprev = NULL;
/*
* Calculate data length.
*/
for (m = m0; m != NULL; m = m->m_next) {
mprev = m;
len += m->m_len;
}
/*
* Make sure packet is actually of even length.
*/
if (len & 1) {
m = mprev;
if ((m->m_flags & M_EXT) == 0 &&
(m->m_len + m->m_data < &m->m_dat[MLEN])) {
mtod(m, char*)[m->m_len++] = 0;
} else {
struct mbuf *m1 = m_get(M_DONTWAIT, MT_DATA);
if (m1 == NULL) {
m_freem(m0);
return (ENOBUFS);
}
m1->m_len = 1;
* mtod(m1, char *) = 0;
m->m_next = m1;
}
m0->m_pkthdr.len++;
}
/*
* Fill in mbuf with extended IPX header
* and addresses and length put into network format.
*/
m = m0;
if (ipxp->ipxp_flags & IPXP_RAWOUT) {
ipx = mtod(m, struct ipx *);
} else {
M_PREPEND(m, sizeof(struct ipx), M_DONTWAIT);
if (m == NULL)
return (ENOBUFS);
ipx = mtod(m, struct ipx *);
ipx->ipx_tc = 0;
ipx->ipx_pt = ipxp->ipxp_dpt;
ipx->ipx_sna = ipxp->ipxp_laddr;
ipx->ipx_dna = ipxp->ipxp_faddr;
len += sizeof(struct ipx);
}
ipx->ipx_len = htons((u_short)len);
if (ipxp->ipxp_flags & IPXP_CHECKSUM) {
ipx->ipx_sum = ipx_cksum(m, len);
} else
ipx->ipx_sum = 0xffff;
/*
* Output datagram.
*/
so = ipxp->ipxp_socket;
if (so->so_options & SO_DONTROUTE)
return (ipx_outputfl(m, (struct route *)NULL,
(so->so_options & SO_BROADCAST) | IPX_ROUTETOIF));
/*
* Use cached route for previous datagram if
* possible. If the previous net was the same
* and the interface was a broadcast medium, or
* if the previous destination was identical,
* then we are ok.
*
* NB: We don't handle broadcasts because that
* would require 3 subroutine calls.
*/
ro = &ipxp->ipxp_route;
#ifdef ancient_history
/*
* I think that this will all be handled in ipx_pcbconnect!
*/
if (ro->ro_rt != NULL) {
if(ipx_neteq(ipxp->ipxp_lastdst, ipx->ipx_dna)) {
/*
* This assumes we have no GH type routes
*/
if (ro->ro_rt->rt_flags & RTF_HOST) {
if (!ipx_hosteq(ipxp->ipxp_lastdst, ipx->ipx_dna))
goto re_route;
}
if ((ro->ro_rt->rt_flags & RTF_GATEWAY) == 0) {
register struct ipx_addr *dst =
&satoipx_addr(ro->ro_dst);
dst->x_host = ipx->ipx_dna.x_host;
}
/*
* Otherwise, we go through the same gateway
* and dst is already set up.
*/
} else {
re_route:
RTFREE(ro->ro_rt);
ro->ro_rt = NULL;
}
}
ipxp->ipxp_lastdst = ipx->ipx_dna;
#endif /* ancient_history */
return (ipx_outputfl(m, ro, so->so_options & SO_BROADCAST));
}
int
ipx_ctloutput(so, sopt)
struct socket *so;
struct sockopt *sopt;
{
struct ipxpcb *ipxp = sotoipxpcb(so);
int mask, error, optval;
short soptval;
struct ipx ioptval;
error = 0;
if (ipxp == NULL)
return (EINVAL);
switch (sopt->sopt_dir) {
case SOPT_GET:
switch (sopt->sopt_name) {
case SO_ALL_PACKETS:
mask = IPXP_ALL_PACKETS;
goto get_flags;
case SO_HEADERS_ON_INPUT:
mask = IPXP_RAWIN;
goto get_flags;
case SO_IPX_CHECKSUM:
mask = IPXP_CHECKSUM;
goto get_flags;
case SO_HEADERS_ON_OUTPUT:
mask = IPXP_RAWOUT;
get_flags:
soptval = ipxp->ipxp_flags & mask;
error = sooptcopyout(sopt, &soptval, sizeof soptval);
break;
case SO_DEFAULT_HEADERS:
ioptval.ipx_len = 0;
ioptval.ipx_sum = 0;
ioptval.ipx_tc = 0;
ioptval.ipx_pt = ipxp->ipxp_dpt;
ioptval.ipx_dna = ipxp->ipxp_faddr;
ioptval.ipx_sna = ipxp->ipxp_laddr;
error = sooptcopyout(sopt, &soptval, sizeof soptval);
break;
case SO_SEQNO:
error = sooptcopyout(sopt, &ipx_pexseq,
sizeof ipx_pexseq);
ipx_pexseq++;
break;
default:
error = EINVAL;
}
break;
case SOPT_SET:
switch (sopt->sopt_name) {
case SO_ALL_PACKETS:
mask = IPXP_ALL_PACKETS;
goto set_head;
case SO_HEADERS_ON_INPUT:
mask = IPXP_RAWIN;
goto set_head;
case SO_IPX_CHECKSUM:
mask = IPXP_CHECKSUM;
case SO_HEADERS_ON_OUTPUT:
mask = IPXP_RAWOUT;
set_head:
error = sooptcopyin(sopt, &optval, sizeof optval,
sizeof optval);
if (error)
break;
if (optval)
ipxp->ipxp_flags |= mask;
else
ipxp->ipxp_flags &= ~mask;
break;
case SO_DEFAULT_HEADERS:
error = sooptcopyin(sopt, &ioptval, sizeof ioptval,
sizeof ioptval);
if (error)
break;
ipxp->ipxp_dpt = ioptval.ipx_pt;
break;
#ifdef IPXIP
case SO_IPXIP_ROUTE:
error = ipxip_route(so, sopt);
break;
#endif /* IPXIP */
default:
error = EINVAL;
}
break;
}
return (error);
}
static int
ipx_usr_abort(so)
struct socket *so;
{
int s;
struct ipxpcb *ipxp = sotoipxpcb(so);
s = splnet();
ipx_pcbdetach(ipxp);
splx(s);
sotryfree(so);
soisdisconnected(so);
return (0);
}
static int
ipx_attach(so, proto, td)
struct socket *so;
int proto;
struct thread *td;
{
int error;
int s;
struct ipxpcb *ipxp = sotoipxpcb(so);
if (ipxp != NULL)
return (EINVAL);
s = splnet();
error = ipx_pcballoc(so, &ipxpcb, td);
splx(s);
if (error == 0)
error = soreserve(so, ipxsendspace, ipxrecvspace);
return (error);
}
static int
ipx_bind(so, nam, td)
struct socket *so;
struct sockaddr *nam;
struct thread *td;
{
struct ipxpcb *ipxp = sotoipxpcb(so);
return (ipx_pcbbind(ipxp, nam, td));
}
static int
ipx_connect(so, nam, td)
struct socket *so;
struct sockaddr *nam;
struct thread *td;
{
int error;
int s;
struct ipxpcb *ipxp = sotoipxpcb(so);
if (!ipx_nullhost(ipxp->ipxp_faddr))
return (EISCONN);
s = splnet();
error = ipx_pcbconnect(ipxp, nam, td);
splx(s);
if (error == 0)
soisconnected(so);
return (error);
}
static int
ipx_detach(so)
struct socket *so;
{
int s;
struct ipxpcb *ipxp = sotoipxpcb(so);
if (ipxp == NULL)
return (ENOTCONN);
s = splnet();
ipx_pcbdetach(ipxp);
splx(s);
return (0);
}
static int
ipx_disconnect(so)
struct socket *so;
{
int s;
struct ipxpcb *ipxp = sotoipxpcb(so);
if (ipx_nullhost(ipxp->ipxp_faddr))
return (ENOTCONN);
s = splnet();
ipx_pcbdisconnect(ipxp);
splx(s);
soisdisconnected(so);
return (0);
}
int
ipx_peeraddr(so, nam)
struct socket *so;
struct sockaddr **nam;
{
struct ipxpcb *ipxp = sotoipxpcb(so);
ipx_setpeeraddr(ipxp, nam); /* XXX what if alloc fails? */
return (0);
}
static int
ipx_send(so, flags, m, nam, control, td)
struct socket *so;
int flags;
struct mbuf *m;
struct sockaddr *nam;
struct mbuf *control;
struct thread *td;
{
int error;
struct ipxpcb *ipxp = sotoipxpcb(so);
struct ipx_addr laddr;
int s = 0;
if (nam != NULL) {
laddr = ipxp->ipxp_laddr;
if (!ipx_nullhost(ipxp->ipxp_faddr)) {
error = EISCONN;
goto send_release;
}
/*
* Must block input while temporarily connected.
*/
s = splnet();
error = ipx_pcbconnect(ipxp, nam, td);
if (error) {
splx(s);
goto send_release;
}
} else {
if (ipx_nullhost(ipxp->ipxp_faddr)) {
error = ENOTCONN;
goto send_release;
}
}
error = ipx_output(ipxp, m);
m = NULL;
if (nam != NULL) {
ipx_pcbdisconnect(ipxp);
splx(s);
ipxp->ipxp_laddr = laddr;
}
send_release:
if (m != NULL)
m_freem(m);
return (error);
}
static int
ipx_shutdown(so)
struct socket *so;
{
socantsendmore(so);
return (0);
}
int
ipx_sockaddr(so, nam)
struct socket *so;
struct sockaddr **nam;
{
struct ipxpcb *ipxp = sotoipxpcb(so);
ipx_setsockaddr(ipxp, nam); /* XXX what if alloc fails? */
return (0);
}
static int
ripx_attach(so, proto, td)
struct socket *so;
int proto;
struct thread *td;
{
int error = 0;
int s;
struct ipxpcb *ipxp = sotoipxpcb(so);
if (td != NULL && (error = suser(td)) != 0)
return (error);
s = splnet();
error = ipx_pcballoc(so, &ipxrawpcb, td);
splx(s);
if (error)
return (error);
error = soreserve(so, ipxsendspace, ipxrecvspace);
if (error)
return (error);
ipxp = sotoipxpcb(so);
ipxp->ipxp_faddr.x_host = ipx_broadhost;
ipxp->ipxp_flags = IPXP_RAWIN | IPXP_RAWOUT;
return (error);
}