NetBSD/sys/netns/idp_usrreq.c

520 lines
10 KiB
C

/* $NetBSD: idp_usrreq.c,v 1.27 2006/05/14 21:20:13 elad Exp $ */
/*
* 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. 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.
*
* @(#)idp_usrreq.c 8.2 (Berkeley) 1/9/95
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: idp_usrreq.c,v 1.27 2006/05/14 21:20:13 elad Exp $");
#include "opt_ns.h" /* NSIP: Xerox NS over IP */
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/errno.h>
#include <sys/stat.h>
#include <sys/proc.h>
#include <sys/kauth.h>
#include <net/if.h>
#include <net/route.h>
#include <netns/ns.h>
#include <netns/ns_pcb.h>
#include <netns/ns_if.h>
#include <netns/ns_var.h>
#include <netns/idp.h>
#include <netns/idp_var.h>
#include <netns/ns_error.h>
#include <machine/stdarg.h>
/*
* IDP protocol implementation.
*/
struct sockaddr_ns idp_ns = { sizeof(idp_ns), AF_NS };
/*
* This may also be called for raw listeners.
*/
void
idp_input(struct mbuf *m, ...)
{
struct nspcb *nsp;
struct idp *idp = mtod(m, struct idp *);
struct ifnet *ifp = m->m_pkthdr.rcvif;
va_list ap;
va_start(ap, m);
nsp = va_arg(ap, struct nspcb *);
va_end(ap);
if (nsp==0)
panic("No nspcb");
/*
* Construct sockaddr format source address.
* Stuff source address and datagram in user buffer.
*/
idp_ns.sns_addr = idp->idp_sna;
if (ns_neteqnn(idp->idp_sna.x_net, ns_zeronet) && ifp) {
struct ifaddr *ifa;
for (ifa = ifp->if_addrlist.tqh_first; ifa != 0;
ifa = ifa->ifa_list.tqe_next) {
if (ifa->ifa_addr->sa_family == AF_NS) {
idp_ns.sns_addr.x_net =
IA_SNS(ifa)->sns_addr.x_net;
break;
}
}
}
nsp->nsp_rpt = idp->idp_pt;
if ( ! (nsp->nsp_flags & NSP_RAWIN) ) {
m->m_len -= sizeof (struct idp);
m->m_pkthdr.len -= sizeof (struct idp);
m->m_data += sizeof (struct idp);
}
if (sbappendaddr(&nsp->nsp_socket->so_rcv, snstosa(&idp_ns), m,
(struct mbuf *)0) == 0)
goto bad;
sorwakeup(nsp->nsp_socket);
return;
bad:
m_freem(m);
}
void
idp_abort(struct nspcb *nsp)
{
struct socket *so = nsp->nsp_socket;
ns_pcbdisconnect(nsp);
soisdisconnected(so);
}
/*
* Drop connection, reporting
* the specified error.
*/
void
idp_drop(struct nspcb *nsp, int errno)
{
struct socket *so = nsp->nsp_socket;
#if 0
/*
* someday, in the xerox world
* we will generate error protocol packets
* announcing that the socket has gone away.
*/
if (TCPS_HAVERCVDSYN(tp->t_state)) {
tp->t_state = TCPS_CLOSED;
(void) tcp_output(tp);
}
#endif
so->so_error = errno;
ns_pcbdisconnect(nsp);
soisdisconnected(so);
}
int
idp_output(struct mbuf *m0, ...)
{
struct nspcb *nsp;
struct mbuf *m;
struct idp *idp;
int len = m0->m_pkthdr.len;
extern int idpcksum;
va_list ap;
va_start(ap, m0);
nsp = va_arg(ap, struct nspcb *);
va_end(ap);
/*
* Make sure packet is actually of even length.
*/
if (len & 1) {
struct mbuf *mprev;
for (mprev = NULL, m = m0; m; m = m->m_next)
mprev = m;
m = mprev;
if ((m->m_flags & M_EXT) == 0 &&
(m->m_len + m->m_data < &m->m_dat[MLEN])) {
m->m_len++;
} else {
struct mbuf *m1 = m_get(M_DONTWAIT, MT_DATA);
if (m1 == 0) {
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 IDP header
* and addresses and length put into network format.
*/
m = m0;
if (nsp->nsp_flags & NSP_RAWOUT) {
idp = mtod(m, struct idp *);
} else {
M_PREPEND(m, sizeof (struct idp), M_DONTWAIT);
if (m == 0)
return (ENOBUFS);
idp = mtod(m, struct idp *);
idp->idp_tc = 0;
idp->idp_pt = nsp->nsp_dpt;
idp->idp_sna = nsp->nsp_laddr;
idp->idp_dna = nsp->nsp_faddr;
len += sizeof (struct idp);
}
idp->idp_len = htons((u_int16_t)len);
if (idpcksum) {
idp->idp_sum = 0;
len = ((len - 1) | 1) + 1;
idp->idp_sum = ns_cksum(m, len);
} else
idp->idp_sum = 0xffff;
/*
* 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.
*/
return (ns_output(m, &nsp->nsp_route,
nsp->nsp_socket->so_options & (SO_DONTROUTE | SO_BROADCAST)));
}
/* ARGSUSED */
int
idp_ctloutput(int req, struct socket *so, int level, int name,
struct mbuf **value)
{
struct mbuf *m;
struct nspcb *nsp = sotonspcb(so);
int mask, error = 0;
if (nsp == NULL)
return (EINVAL);
switch (req) {
case PRCO_GETOPT:
if (value==NULL)
return (EINVAL);
m = m_get(M_DONTWAIT, MT_DATA);
if (m==NULL)
return (ENOBUFS);
switch (name) {
case SO_ALL_PACKETS:
mask = NSP_ALL_PACKETS;
goto get_flags;
case SO_HEADERS_ON_INPUT:
mask = NSP_RAWIN;
goto get_flags;
case SO_HEADERS_ON_OUTPUT:
mask = NSP_RAWOUT;
get_flags:
m->m_len = sizeof(short);
*mtod(m, short *) = nsp->nsp_flags & mask;
break;
case SO_DEFAULT_HEADERS:
m->m_len = sizeof(struct idp);
{
struct idp *idp = mtod(m, struct idp *);
idp->idp_len = 0;
idp->idp_sum = 0;
idp->idp_tc = 0;
idp->idp_pt = nsp->nsp_dpt;
idp->idp_dna = nsp->nsp_faddr;
idp->idp_sna = nsp->nsp_laddr;
}
break;
case SO_SEQNO:
m->m_len = sizeof(long);
*mtod(m, long *) = ns_pexseq++;
break;
default:
error = EINVAL;
}
*value = m;
break;
case PRCO_SETOPT:
switch (name) {
int *ok;
case SO_ALL_PACKETS:
mask = NSP_ALL_PACKETS;
goto set_head;
case SO_HEADERS_ON_INPUT:
mask = NSP_RAWIN;
goto set_head;
case SO_HEADERS_ON_OUTPUT:
mask = NSP_RAWOUT;
set_head:
if (value && *value) {
ok = mtod(*value, int *);
if (*ok)
nsp->nsp_flags |= mask;
else
nsp->nsp_flags &= ~mask;
} else error = EINVAL;
break;
case SO_DEFAULT_HEADERS:
{
struct idp *idp
= mtod(*value, struct idp *);
nsp->nsp_dpt = idp->idp_pt;
}
break;
#ifdef NSIP
case SO_NSIP_ROUTE:
error = nsip_route(*value);
break;
#endif /* NSIP */
default:
error = EINVAL;
}
if (value && *value)
m_freem(*value);
break;
}
return (error);
}
u_long idp_sendspace = 2048;
u_long idp_recvspace = 2048;
/*ARGSUSED*/
int
idp_usrreq(struct socket *so, int req, struct mbuf *m, struct mbuf *nam,
struct mbuf *control, struct lwp *l)
{
struct nspcb *nsp;
struct proc *p;
int s;
int error = 0;
p = l ? l->l_proc : NULL;
if (req == PRU_CONTROL)
return (ns_control(so, (u_long)m, (caddr_t)nam,
(struct ifnet *)control, p));
if (req == PRU_PURGEIF) {
ns_purgeif((struct ifnet *)control);
return (0);
}
s = splsoftnet();
nsp = sotonspcb(so);
if (nsp == 0 && req != PRU_ATTACH) {
error = EINVAL;
goto release;
}
switch (req) {
case PRU_ATTACH:
if (nsp != 0) {
error = EISCONN;
break;
}
if ((error = soreserve(so, idp_sendspace, idp_recvspace)) ||
(error = ns_pcballoc(so, &nspcb)))
break;
break;
case PRU_DETACH:
ns_pcbdetach(nsp);
break;
case PRU_BIND:
error = ns_pcbbind(nsp, nam, p);
break;
case PRU_LISTEN:
error = EOPNOTSUPP;
break;
case PRU_CONNECT:
error = ns_pcbconnect(nsp, nam);
if (error)
break;
soisconnected(so);
break;
case PRU_CONNECT2:
error = EOPNOTSUPP;
break;
case PRU_DISCONNECT:
soisdisconnected(so);
ns_pcbdisconnect(nsp);
break;
case PRU_SHUTDOWN:
socantsendmore(so);
break;
case PRU_RCVD:
error = EOPNOTSUPP;
break;
case PRU_SEND:
{
struct ns_addr laddr;
if (nam) {
laddr = nsp->nsp_laddr;
if ((so->so_state & SS_ISCONNECTED) != 0) {
error = EISCONN;
break;
}
error = ns_pcbconnect(nsp, nam);
if (error)
break;
} else {
if ((so->so_state & SS_ISCONNECTED) == 0) {
error = ENOTCONN;
break;
}
}
error = idp_output(m, nsp);
if (nam) {
ns_pcbdisconnect(nsp);
nsp->nsp_laddr = laddr;
}
}
break;
case PRU_SENSE:
/*
* stat: don't bother with a blocksize.
*/
splx(s);
return (0);
/*
* Not supported.
*/
case PRU_RCVOOB:
error = EOPNOTSUPP;
break;
case PRU_SENDOOB:
m_freem(m);
error = EOPNOTSUPP;
break;
case PRU_SOCKADDR:
ns_setsockaddr(nsp, nam);
break;
case PRU_PEERADDR:
ns_setpeeraddr(nsp, nam);
break;
default:
panic("idp_usrreq");
}
release:
splx(s);
return (error);
}
/*ARGSUSED*/
int
idp_raw_usrreq(struct socket *so, int req, struct mbuf *m, struct mbuf *nam,
struct mbuf *control, struct lwp *l)
{
int error = 0;
struct proc *p;
struct nspcb *nsp = sotonspcb(so);
p = l->l_proc;
switch (req) {
case PRU_ATTACH:
if (nsp != 0) {
error = EISCONN;
break;
}
if (p == 0 || (error = kauth_authorize_generic(p->p_cred, KAUTH_GENERIC_ISSUSER,
&p->p_acflag))) {
error = EACCES;
break;
}
if ((error = soreserve(so, idp_sendspace, idp_recvspace)) ||
(error = ns_pcballoc(so, &nspcb)))
break;
nsp = sotonspcb(so);
nsp->nsp_faddr.x_host = ns_broadhost;
nsp->nsp_flags = NSP_RAWIN | NSP_RAWOUT;
break;
default:
error = idp_usrreq(so, req, m, nam, control, l);
break;
}
return (error);
}