Please note that diffs are not public domain; they are subject to the copyright notices on the relevant files. =================================================================== RCS file: /ftp/cvs/cvsroot/src/sys/netinet/raw_ip.c,v rcsdiff: /ftp/cvs/cvsroot/src/sys/netinet/raw_ip.c,v: warning: Unknown phrases like `commitid ...;' are present. retrieving revision 1.2 retrieving revision 1.43 diff -u -p -r1.2 -r1.43 --- src/sys/netinet/raw_ip.c 1993/03/21 18:04:42 1.2 +++ src/sys/netinet/raw_ip.c 1999/07/01 08:12:51 1.43 @@ -1,6 +1,37 @@ +/* $NetBSD: raw_ip.c,v 1.43 1999/07/01 08:12:51 itojun Exp $ */ + /* - * Copyright (c) 1982, 1986, 1988 Regents of the University of California. + * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. * 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 project 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 PROJECT 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 PROJECT 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. + */ + +/* + * Copyright (c) 1982, 1986, 1988, 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 @@ -30,240 +61,513 @@ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * - * @(#)raw_ip.c 7.8 (Berkeley) 7/25/90 - * - * PATCHES MAGIC LEVEL PATCH THAT GOT US HERE - * -------------------- ----- ---------------------- - * CURRENT PATCH LEVEL: 1 00020 - * -------------------- ----- ---------------------- - * - * 03 Nov 92 Julian Elischer Fixed memory leak that caused ping - * and traceroute to cause panic + * @(#)raw_ip.c 8.7 (Berkeley) 5/15/95 */ -#include "param.h" -#include "malloc.h" -#include "mbuf.h" -#include "socket.h" -#include "protosw.h" -#include "socketvar.h" -#include "errno.h" - -#include "../net/if.h" -#include "../net/route.h" -#include "../net/raw_cb.h" - -#include "in.h" -#include "in_systm.h" -#include "ip.h" -#include "ip_var.h" -#include "in_pcb.h" +#include "opt_mrouting.h" + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include + +#include +#include +#include +#include +#include +#include +#include + +#include + +#ifdef IPSEC +#include +#endif /*IPSEC*/ + +struct inpcbtable rawcbtable; + +int rip_bind __P((struct inpcb *, struct mbuf *)); +int rip_connect __P((struct inpcb *, struct mbuf *)); +void rip_disconnect __P((struct inpcb *)); + +/* + * Nominal space allocated to a raw ip socket. + */ +#define RIPSNDQ 8192 +#define RIPRCVQ 8192 /* * Raw interface to IP protocol. */ -struct sockaddr_in ripdst = { sizeof(ripdst), AF_INET }; -struct sockaddr_in ripsrc = { sizeof(ripsrc), AF_INET }; -struct sockproto ripproto = { PF_INET }; +/* + * Initialize raw connection block q. + */ +void +rip_init() +{ + + in_pcbinit(&rawcbtable, 1, 1); +} + +static struct sockaddr_in ripsrc = { sizeof(ripsrc), AF_INET }; + /* * Setup generic address and protocol structures * for raw_input routine, then pass them along with * mbuf chain. */ -rip_input(m) +void +#if __STDC__ +rip_input(struct mbuf *m, ...) +#else +rip_input(m, va_alist) struct mbuf *m; + va_dcl +#endif { + int off, proto; register struct ip *ip = mtod(m, struct ip *); + register struct inpcb *inp; + struct inpcb *last = 0; + struct mbuf *opts = 0; + struct sockaddr_in ripsrc; + va_list ap; + + va_start(ap, m); + off = va_arg(ap, int); + proto = va_arg(ap, int); + va_end(ap); - ripproto.sp_protocol = ip->ip_p; - ripdst.sin_addr = ip->ip_dst; + ripsrc.sin_family = AF_INET; + ripsrc.sin_len = sizeof(struct sockaddr_in); ripsrc.sin_addr = ip->ip_src; - if (raw_input(m, &ripproto, (struct sockaddr *)&ripsrc, - (struct sockaddr *)&ripdst) == 0) { + ripsrc.sin_port = 0; + bzero((caddr_t)ripsrc.sin_zero, sizeof(ripsrc.sin_zero)); + + /* + * XXX Compatibility: programs using raw IP expect ip_len + * XXX to have the header length subtracted. + */ + ip->ip_len -= ip->ip_hl << 2; + + for (inp = rawcbtable.inpt_queue.cqh_first; + inp != (struct inpcb *)&rawcbtable.inpt_queue; + inp = inp->inp_queue.cqe_next) { + if (inp->inp_ip.ip_p && inp->inp_ip.ip_p != proto) + continue; + if (!in_nullhost(inp->inp_laddr) && + !in_hosteq(inp->inp_laddr, ip->ip_dst)) + continue; + if (!in_nullhost(inp->inp_faddr) && + !in_hosteq(inp->inp_faddr, ip->ip_src)) + continue; + if (last) { + struct mbuf *n; + if ((n = m_copy(m, 0, (int)M_COPYALL)) != NULL) { + if (last->inp_flags & INP_CONTROLOPTS || + last->inp_socket->so_options & SO_TIMESTAMP) + ip_savecontrol(last, &opts, ip, n); + if (sbappendaddr(&last->inp_socket->so_rcv, + sintosa(&ripsrc), n, opts) == 0) { + /* should notify about lost packet */ + m_freem(n); + if (opts) + m_freem(opts); + } else + sorwakeup(last->inp_socket); + opts = NULL; + } + } + last = inp; + } + if (last) { + if (last->inp_flags & INP_CONTROLOPTS || + last->inp_socket->so_options & SO_TIMESTAMP) + ip_savecontrol(last, &opts, ip, m); + if (sbappendaddr(&last->inp_socket->so_rcv, + sintosa(&ripsrc), m, opts) == 0) { + m_freem(m); + if (opts) + m_freem(opts); + } else + sorwakeup(last->inp_socket); + } else { + m_freem(m); ipstat.ips_noproto++; ipstat.ips_delivered--; } + return; } /* * Generate IP header and pass packet to ip_output. * Tack on options user may have setup with control call. */ -#define satosin(sa) ((struct sockaddr_in *)(sa)) -rip_output(m, so) - register struct mbuf *m; - struct socket *so; +int +#if __STDC__ +rip_output(struct mbuf *m, ...) +#else +rip_output(m, va_alist) + struct mbuf *m; + va_dcl +#endif { + register struct inpcb *inp; register struct ip *ip; - register struct raw_inpcb *rp = sotorawinpcb(so); - register struct sockaddr_in *sin; + struct mbuf *opts; + int flags; + va_list ap; + + va_start(ap, m); + inp = va_arg(ap, struct inpcb *); + va_end(ap); + + flags = + (inp->inp_socket->so_options & SO_DONTROUTE) | IP_ALLOWBROADCAST + | IP_RETURNMTU; /* * If the user handed us a complete IP packet, use it. * Otherwise, allocate an mbuf for a header and fill it in. */ - if (rp->rinp_flags & RINPF_HDRINCL) - ip = mtod(m, struct ip *); - else { + if ((inp->inp_flags & INP_HDRINCL) == 0) { + if ((m->m_pkthdr.len + sizeof(struct ip)) > IP_MAXPACKET) { + m_freem(m); + return (EMSGSIZE); + } M_PREPEND(m, sizeof(struct ip), M_WAIT); ip = mtod(m, struct ip *); ip->ip_tos = 0; ip->ip_off = 0; - ip->ip_p = rp->rinp_rcb.rcb_proto.sp_protocol; + ip->ip_p = inp->inp_ip.ip_p; ip->ip_len = m->m_pkthdr.len; - if (sin = satosin(rp->rinp_rcb.rcb_laddr)) { - ip->ip_src = sin->sin_addr; - } else - ip->ip_src.s_addr = 0; - if (sin = satosin(rp->rinp_rcb.rcb_faddr)) - ip->ip_dst = sin->sin_addr; + ip->ip_src = inp->inp_laddr; + ip->ip_dst = inp->inp_faddr; ip->ip_ttl = MAXTTL; + opts = inp->inp_options; + } else { + if (m->m_pkthdr.len > IP_MAXPACKET) { + m_freem(m); + return (EMSGSIZE); + } + ip = mtod(m, struct ip *); + if (m->m_pkthdr.len != ip->ip_len) { + m_freem(m); + return (EINVAL); + } + if (ip->ip_id == 0) + ip->ip_id = htons(ip_id++); + opts = NULL; + /* XXX prevent ip_output from overwriting header fields */ + flags |= IP_RAWOUTPUT; + ipstat.ips_rawout++; } - return (ip_output(m, - (rp->rinp_flags & RINPF_HDRINCL)? (struct mbuf *)0: rp->rinp_options, - &rp->rinp_route, - (so->so_options & SO_DONTROUTE) | IP_ALLOWBROADCAST)); +#ifdef IPSEC + m->m_pkthdr.rcvif = (struct ifnet *)inp->inp_socket; /*XXX*/ +#endif /*IPSEC*/ + return (ip_output(m, opts, &inp->inp_route, flags, inp->inp_moptions, &inp->inp_errormtu)); } /* * Raw IP socket option processing. */ +int rip_ctloutput(op, so, level, optname, m) int op; struct socket *so; int level, optname; struct mbuf **m; { + register struct inpcb *inp = sotoinpcb(so); int error = 0; - register struct raw_inpcb *rp = sotorawinpcb(so); - if (level != IPPROTO_IP) - error = EINVAL; - else switch (op) { + if (level != IPPROTO_IP) { + error = ENOPROTOOPT; + if (op == PRCO_SETOPT && *m != 0) + (void) m_free(*m); + } else switch (op) { case PRCO_SETOPT: switch (optname) { - - case IP_OPTIONS: - return (ip_pcbopts(&rp->rinp_options, *m)); - case IP_HDRINCL: - if (m == 0 || *m == 0 || (*m)->m_len < sizeof (int)) { + if (*m == 0 || (*m)->m_len < sizeof (int)) error = EINVAL; - break; + else { + if (*mtod(*m, int *)) + inp->inp_flags |= INP_HDRINCL; + else + inp->inp_flags &= ~INP_HDRINCL; } - if (*mtod(*m, int *)) - rp->rinp_flags |= RINPF_HDRINCL; - else - rp->rinp_flags &= ~RINPF_HDRINCL; + if (*m != 0) + (void) m_free(*m); break; +#ifdef MROUTING + case MRT_INIT: + case MRT_DONE: + case MRT_ADD_VIF: + case MRT_DEL_VIF: + case MRT_ADD_MFC: + case MRT_DEL_MFC: + case MRT_ASSERT: + error = ip_mrouter_set(so, optname, m); + break; +#endif + default: - error = EINVAL; + error = ip_ctloutput(op, so, level, optname, m); break; } break; case PRCO_GETOPT: - *m = m_get(M_WAIT, MT_SOOPTS); switch (optname) { - - case IP_OPTIONS: - if (rp->rinp_options) { - (*m)->m_len = rp->rinp_options->m_len; - bcopy(mtod(rp->rinp_options, caddr_t), - mtod(*m, caddr_t), (unsigned)(*m)->m_len); - } else - (*m)->m_len = 0; - break; - case IP_HDRINCL: + *m = m_get(M_WAIT, M_SOOPTS); (*m)->m_len = sizeof (int); - *mtod(*m, int *) = rp->rinp_flags & RINPF_HDRINCL; + *mtod(*m, int *) = inp->inp_flags & INP_HDRINCL ? 1 : 0; + break; + +#ifdef MROUTING + case MRT_VERSION: + case MRT_ASSERT: + error = ip_mrouter_get(so, optname, m); break; +#endif default: - error = EINVAL; - m_freem(*m); - *m = 0; + error = ip_ctloutput(op, so, level, optname, m); break; } break; } - if (op == PRCO_SETOPT && *m) - (void)m_free(*m); return (error); } +int +rip_bind(inp, nam) + struct inpcb *inp; + struct mbuf *nam; +{ + struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); + + if (nam->m_len != sizeof(*addr)) + return (EINVAL); + if (ifnet.tqh_first == 0) + return (EADDRNOTAVAIL); + if (addr->sin_family != AF_INET && + addr->sin_family != AF_IMPLINK) + return (EAFNOSUPPORT); + if (!in_nullhost(addr->sin_addr) && + ifa_ifwithaddr(sintosa(addr)) == 0) + return (EADDRNOTAVAIL); + inp->inp_laddr = addr->sin_addr; + return (0); +} + +int +rip_connect(inp, nam) + struct inpcb *inp; + struct mbuf *nam; +{ + struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); + + if (nam->m_len != sizeof(*addr)) + return (EINVAL); + if (ifnet.tqh_first == 0) + return (EADDRNOTAVAIL); + if (addr->sin_family != AF_INET && + addr->sin_family != AF_IMPLINK) + return (EAFNOSUPPORT); + inp->inp_faddr = addr->sin_addr; + return (0); +} + +void +rip_disconnect(inp) + struct inpcb *inp; +{ + + inp->inp_faddr = zeroin_addr; +} + +u_long rip_sendspace = RIPSNDQ; +u_long rip_recvspace = RIPRCVQ; + /*ARGSUSED*/ -rip_usrreq(so, req, m, nam, control) +int +rip_usrreq(so, req, m, nam, control, p) register struct socket *so; int req; struct mbuf *m, *nam, *control; + struct proc *p; { + register struct inpcb *inp; + int s; register int error = 0; - register struct raw_inpcb *rp = sotorawinpcb(so); +#ifdef MROUTING + extern struct socket *ip_mrouter; +#endif + + if (req == PRU_CONTROL) + return (in_control(so, (long)m, (caddr_t)nam, + (struct ifnet *)control, p)); + + s = splsoftnet(); + inp = sotoinpcb(so); +#ifdef DIAGNOSTIC + if (req != PRU_SEND && req != PRU_SENDOOB && control) + panic("rip_usrreq: unexpected control mbuf"); +#endif + if (inp == 0 && req != PRU_ATTACH) { + error = EINVAL; + goto release; + } switch (req) { case PRU_ATTACH: - if (rp) - panic("rip_attach"); - MALLOC(rp, struct raw_inpcb *, sizeof *rp, M_PCB, M_WAITOK); - if (rp == 0) - return (ENOBUFS); - bzero((caddr_t)rp, sizeof *rp); - so->so_pcb = (caddr_t)rp; + if (inp != 0) { + error = EISCONN; + break; + } + if (p == 0 || (error = suser(p->p_ucred, &p->p_acflag))) { + error = EACCES; + break; + } + if (so->so_snd.sb_hiwat == 0 || so->so_rcv.sb_hiwat == 0) { + error = soreserve(so, rip_sendspace, rip_recvspace); + if (error) + break; + } + error = in_pcballoc(so, &rawcbtable); + if (error) + break; + inp = sotoinpcb(so); + inp->inp_ip.ip_p = (long)nam; +#ifdef IPSEC + error = ipsec_init_policy(&inp->inp_sp); +#endif /*IPSEC*/ break; case PRU_DETACH: - if (rp == 0) - panic("rip_detach"); - if (rp->rinp_options) - m_freem(rp->rinp_options); - if (rp->rinp_route.ro_rt) - RTFREE(rp->rinp_route.ro_rt); - if (rp->rinp_rcb.rcb_laddr) - rp->rinp_rcb.rcb_laddr = 0; +#ifdef MROUTING + if (so == ip_mrouter) + ip_mrouter_done(); +#endif + in_pcbdetach(inp); break; case PRU_BIND: - { - struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); + error = rip_bind(inp, nam); + break; - if (nam->m_len != sizeof(*addr)) - return (EINVAL); - if ((ifnet == 0) || - ((addr->sin_family != AF_INET) && - (addr->sin_family != AF_IMPLINK)) || - (addr->sin_addr.s_addr && - ifa_ifwithaddr((struct sockaddr *)addr) == 0)) - return (EADDRNOTAVAIL); - rp->rinp_rcb.rcb_laddr = (struct sockaddr *)&rp->rinp_laddr; - rp->rinp_laddr = *addr; - return (0); - } - case PRU_CONNECT: - { - struct sockaddr_in *addr = mtod(nam, struct sockaddr_in *); + case PRU_LISTEN: + error = EOPNOTSUPP; + break; - if (nam->m_len != sizeof(*addr)) - return (EINVAL); - if (ifnet == 0) - return (EADDRNOTAVAIL); - if ((addr->sin_family != AF_INET) && - (addr->sin_family != AF_IMPLINK)) - return (EAFNOSUPPORT); - rp->rinp_rcb.rcb_faddr = (struct sockaddr *)&rp->rinp_faddr; - rp->rinp_faddr = *addr; + case PRU_CONNECT: + error = rip_connect(inp, nam); + if (error) + break; soisconnected(so); + break; + + case PRU_CONNECT2: + error = EOPNOTSUPP; + break; + + case PRU_DISCONNECT: + soisdisconnected(so); + rip_disconnect(inp); + break; + + /* + * Mark the connection as being incapable of further input. + */ + case PRU_SHUTDOWN: + socantsendmore(so); + break; + + case PRU_RCVD: + error = EOPNOTSUPP; + break; + + /* + * Ship a packet out. The appropriate raw output + * routine handles any massaging necessary. + */ + case PRU_SEND: + if (control && control->m_len) { + m_freem(control); + m_freem(m); + error = EINVAL; + break; + } + { + if (nam) { + if ((so->so_state & SS_ISCONNECTED) != 0) { + error = EISCONN; + goto die; + } + error = rip_connect(inp, nam); + if (error) { + die: + m_freem(m); + break; + } + } else { + if ((so->so_state & SS_ISCONNECTED) == 0) { + error = ENOTCONN; + goto die; + } + } + error = rip_output(m, inp); + if (nam) + rip_disconnect(inp); + } + break; + + case PRU_SENSE: + /* + * stat: don't bother with a blocksize. + */ + splx(s); return (0); - } + + case PRU_RCVOOB: + error = EOPNOTSUPP; + break; + + case PRU_SENDOOB: + m_freem(control); + m_freem(m); + error = EOPNOTSUPP; + break; + + case PRU_SOCKADDR: + in_setsockaddr(inp, nam); + break; + + case PRU_PEERADDR: + in_setpeeraddr(inp, nam); + break; + + default: + panic("rip_usrreq"); } - error = raw_usrreq(so, req, m, nam, control); - if (error && (req == PRU_ATTACH) && so->so_pcb) - free(so->so_pcb, M_PCB); +release: + splx(s); return (error); }