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 1.44 retrieving revision 1.88.4.2 diff -u -p -r1.44 -r1.88.4.2 --- src/sys/netinet/raw_ip.c 1999/07/05 07:24:38 1.44 +++ src/sys/netinet/raw_ip.c 2006/09/09 02:58:47 1.88.4.2 @@ -1,9 +1,9 @@ -/* $NetBSD: raw_ip.c,v 1.44 1999/07/05 07:24:38 darrenr Exp $ */ +/* $NetBSD: raw_ip.c,v 1.88.4.2 2006/09/09 02:58:47 rpaulo Exp $ */ /* * 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: @@ -15,7 +15,7 @@ * 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 @@ -41,11 +41,7 @@ * 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 + * 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. * @@ -64,9 +60,15 @@ * @(#)raw_ip.c 8.7 (Berkeley) 5/15/95 */ +#include +__KERNEL_RCSID(0, "$NetBSD: raw_ip.c,v 1.88.4.2 2006/09/09 02:58:47 rpaulo Exp $"); + +#include "opt_inet.h" +#include "opt_ipsec.h" #include "opt_mrouting.h" #include +#include #include #include #include @@ -75,6 +77,7 @@ #include #include #include +#include #include #include @@ -86,6 +89,7 @@ #include #include #include +#include #include #include @@ -94,13 +98,18 @@ #include #endif /*IPSEC*/ -extern u_char ip_protox[]; -extern struct protosw inetsw[]; +#ifdef FAST_IPSEC +#include +#include /* XXX ipsecstat namespace */ +#endif /* FAST_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 *)); +int rip_pcbnotify(struct inpcbtable *, struct in_addr, + struct in_addr, int, int, void (*)(struct inpcb *, int)); +int rip_bind(struct inpcb *, struct mbuf *); +int rip_connect(struct inpcb *, struct mbuf *); +void rip_disconnect(struct inpcb *); /* * Nominal space allocated to a raw ip socket. @@ -116,38 +125,30 @@ void rip_disconnect __P((struct inpcb * * Initialize raw connection block q. */ void -rip_init() +rip_init(void) { 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. */ 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; + int proto; + struct ip *ip = mtod(m, struct ip *); + 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); + (void)va_arg(ap, int); /* ignore value, advance ap */ proto = va_arg(ap, int); va_end(ap); @@ -159,13 +160,15 @@ rip_input(m, va_alist) /* * XXX Compatibility: programs using raw IP expect ip_len - * XXX to have the header length subtracted. + * XXX to have the header length subtracted, and in host order. + * XXX ip_off is also expected to be host order. */ - ip->ip_len -= ip->ip_hl << 2; + ip->ip_len = ntohs(ip->ip_len) - (ip->ip_hl << 2); + NTOHS(ip->ip_off); - for (inp = rawcbtable.inpt_queue.cqh_first; - inp != (struct inpcb *)&rawcbtable.inpt_queue; - inp = inp->inp_queue.cqe_next) { + CIRCLEQ_FOREACH(inp, &rawcbtable.inpt_queue, inp_queue) { + if (inp->inp_af != AF_INET) + continue; if (inp->inp_ip.ip_p && inp->inp_ip.ip_p != proto) continue; if (!in_nullhost(inp->inp_laddr) && @@ -176,6 +179,14 @@ rip_input(m, va_alist) continue; if (last) { struct mbuf *n; + +#if defined(IPSEC) || defined(FAST_IPSEC) + /* check AH/ESP integrity. */ + if (ipsec4_in_reject_so(m, last->inp_socket)) { + ipsecstat.in_polvio++; + /* do not inject data to pcb */ + } else +#endif /*IPSEC*/ if ((n = m_copy(m, 0, (int)M_COPYALL)) != NULL) { if (last->inp_flags & INP_CONTROLOPTS || last->inp_socket->so_options & SO_TIMESTAMP) @@ -193,6 +204,15 @@ rip_input(m, va_alist) } last = inp; } +#if defined(IPSEC) || defined(FAST_IPSEC) + /* check AH/ESP integrity. */ + if (last && ipsec4_in_reject_so(m, last->inp_socket)) { + m_freem(m); + ipsecstat.in_polvio++; + ipstat.ips_delivered--; + /* do not inject data to pcb */ + } else +#endif /*IPSEC*/ if (last) { if (last->inp_flags & INP_CONTROLOPTS || last->inp_socket->so_options & SO_TIMESTAMP) @@ -206,7 +226,8 @@ rip_input(m, va_alist) sorwakeup(last->inp_socket); } else { if (inetsw[ip_protox[ip->ip_p]].pr_input == rip_input) { - icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_PROTOCOL,0,0); + icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_PROTOCOL, + 0, 0); ipstat.ips_noproto++; ipstat.ips_delivered--; } else @@ -215,21 +236,72 @@ rip_input(m, va_alist) return; } +int +rip_pcbnotify(struct inpcbtable *table, + struct in_addr faddr, struct in_addr laddr, int proto, int errno, + void (*notify)(struct inpcb *, int)) +{ + struct inpcb *inp, *ninp; + int nmatch; + + nmatch = 0; + for (inp = (struct inpcb *)CIRCLEQ_FIRST(&table->inpt_queue); + inp != (struct inpcb *)&table->inpt_queue; + inp = ninp) { + ninp = (struct inpcb *)inp->inp_queue.cqe_next; + if (inp->inp_af != AF_INET) + continue; + if (inp->inp_ip.ip_p && inp->inp_ip.ip_p != proto) + continue; + if (in_hosteq(inp->inp_faddr, faddr) && + in_hosteq(inp->inp_laddr, laddr)) { + (*notify)(inp, errno); + nmatch++; + } + } + + return nmatch; +} + +void * +rip_ctlinput(int cmd, struct sockaddr *sa, void *v) +{ + struct ip *ip = v; + void (*notify)(struct inpcb *, int) = in_rtchange; + int errno; + + if (sa->sa_family != AF_INET || + sa->sa_len != sizeof(struct sockaddr_in)) + return NULL; + if ((unsigned)cmd >= PRC_NCMDS) + return NULL; + errno = inetctlerrmap[cmd]; + if (PRC_IS_REDIRECT(cmd)) + notify = in_rtchange, ip = 0; + else if (cmd == PRC_HOSTDEAD) + ip = 0; + else if (errno == 0) + return NULL; + if (ip) { + rip_pcbnotify(&rawcbtable, satosin(sa)->sin_addr, + ip->ip_src, ip->ip_p, errno, notify); + + /* XXX mapped address case */ + } else + in_pcbnotifyall(&rawcbtable, satosin(sa)->sin_addr, errno, + notify); + return NULL; +} + /* * Generate IP header and pass packet to ip_output. * Tack on options user may have setup with control call. */ 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; + struct inpcb *inp; + struct ip *ip; struct mbuf *opts; int flags; va_list ap; @@ -251,12 +323,14 @@ rip_output(m, va_alist) m_freem(m); return (EMSGSIZE); } - M_PREPEND(m, sizeof(struct ip), M_WAIT); + M_PREPEND(m, sizeof(struct ip), M_DONTWAIT); + if (!m) + return (ENOBUFS); ip = mtod(m, struct ip *); ip->ip_tos = 0; - ip->ip_off = 0; + ip->ip_off = htons(0); ip->ip_p = inp->inp_ip.ip_p; - ip->ip_len = m->m_pkthdr.len; + ip->ip_len = htons(m->m_pkthdr.len); ip->ip_src = inp->inp_laddr; ip->ip_dst = inp->inp_faddr; ip->ip_ttl = MAXTTL; @@ -267,34 +341,47 @@ rip_output(m, va_alist) return (EMSGSIZE); } ip = mtod(m, struct ip *); + + /* + * If the mbuf is read-only, we need to allocate + * a new mbuf for the header, since we need to + * modify the header. + */ + if (M_READONLY(m)) { + int hlen = ip->ip_hl << 2; + + m = m_copyup(m, hlen, (max_linkhdr + 3) & ~3); + if (m == NULL) + return (ENOMEM); /* XXX */ + ip = mtod(m, struct ip *); + } + + /* XXX userland passes ip_len and ip_off in host order */ if (m->m_pkthdr.len != ip->ip_len) { m_freem(m); return (EINVAL); } + HTONS(ip->ip_len); + HTONS(ip->ip_off); if (ip->ip_id == 0) - ip->ip_id = htons(ip_id++); + ip->ip_id = ip_newid(); opts = NULL; /* XXX prevent ip_output from overwriting header fields */ flags |= IP_RAWOUTPUT; ipstat.ips_rawout++; } -#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)); + return (ip_output(m, opts, &inp->inp_route, flags, inp->inp_moptions, + inp->inp_socket, &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; +rip_ctloutput(int op, struct socket *so, int level, int optname, + struct mbuf **m) { - register struct inpcb *inp = sotoinpcb(so); + struct inpcb *inp = sotoinpcb(so); int error = 0; if (level != IPPROTO_IP) { @@ -326,6 +413,9 @@ rip_ctloutput(op, so, level, optname, m) case MRT_ADD_MFC: case MRT_DEL_MFC: case MRT_ASSERT: + case MRT_API_CONFIG: + case MRT_ADD_BW_UPCALL: + case MRT_DEL_BW_UPCALL: error = ip_mrouter_set(so, optname, m); break; #endif @@ -339,7 +429,8 @@ rip_ctloutput(op, so, level, optname, m) case PRCO_GETOPT: switch (optname) { case IP_HDRINCL: - *m = m_get(M_WAIT, M_SOOPTS); + *m = m_get(M_WAIT, MT_SOOPTS); + MCLAIM((*m), so->so_mowner); (*m)->m_len = sizeof (int); *mtod(*m, int *) = inp->inp_flags & INP_HDRINCL ? 1 : 0; break; @@ -347,6 +438,8 @@ rip_ctloutput(op, so, level, optname, m) #ifdef MROUTING case MRT_VERSION: case MRT_ASSERT: + case MRT_API_SUPPORT: + case MRT_API_CONFIG: error = ip_mrouter_get(so, optname, m); break; #endif @@ -361,15 +454,13 @@ rip_ctloutput(op, so, level, optname, m) } int -rip_bind(inp, nam) - struct inpcb *inp; - struct mbuf *nam; +rip_bind(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) + if (TAILQ_FIRST(&ifnet) == 0) return (EADDRNOTAVAIL); if (addr->sin_family != AF_INET && addr->sin_family != AF_IMPLINK) @@ -382,15 +473,13 @@ rip_bind(inp, nam) } int -rip_connect(inp, nam) - struct inpcb *inp; - struct mbuf *nam; +rip_connect(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) + if (TAILQ_FIRST(&ifnet) == 0) return (EADDRNOTAVAIL); if (addr->sin_family != AF_INET && addr->sin_family != AF_IMPLINK) @@ -400,8 +489,7 @@ rip_connect(inp, nam) } void -rip_disconnect(inp) - struct inpcb *inp; +rip_disconnect(struct inpcb *inp) { inp->inp_faddr = zeroin_addr; @@ -412,22 +500,26 @@ u_long rip_recvspace = RIPRCVQ; /*ARGSUSED*/ int -rip_usrreq(so, req, m, nam, control, p) - register struct socket *so; - int req; - struct mbuf *m, *nam, *control; - struct proc *p; +rip_usrreq(struct socket *so, int req, + struct mbuf *m, struct mbuf *nam, struct mbuf *control, struct lwp *l) { - register struct inpcb *inp; + struct inpcb *inp; int s; - register int error = 0; + int error = 0; #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)); + (struct ifnet *)control, l)); + + if (req == PRU_PURGEIF) { + in_pcbpurgeif0(&rawcbtable, (struct ifnet *)control); + in_purgeif((struct ifnet *)control); + in_pcbpurgeif(&rawcbtable, (struct ifnet *)control); + return (0); + } s = splsoftnet(); inp = sotoinpcb(so); @@ -447,7 +539,8 @@ rip_usrreq(so, req, m, nam, control, p) error = EISCONN; break; } - if (p == 0 || (error = suser(p->p_ucred, &p->p_acflag))) { + if (l == 0 || (error = kauth_authorize_generic(l->l_cred, + KAUTH_GENERIC_ISSUSER, &l->l_acflag))) { error = EACCES; break; } @@ -461,9 +554,6 @@ rip_usrreq(so, req, m, nam, control, p) 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: @@ -577,3 +667,32 @@ release: splx(s); return (error); } + +SYSCTL_SETUP(sysctl_net_inet_raw_setup, "sysctl net.inet.raw subtree setup") +{ + + sysctl_createv(clog, 0, NULL, NULL, + CTLFLAG_PERMANENT, + CTLTYPE_NODE, "net", NULL, + NULL, 0, NULL, 0, + CTL_NET, CTL_EOL); + sysctl_createv(clog, 0, NULL, NULL, + CTLFLAG_PERMANENT, + CTLTYPE_NODE, "inet", NULL, + NULL, 0, NULL, 0, + CTL_NET, PF_INET, CTL_EOL); + sysctl_createv(clog, 0, NULL, NULL, + CTLFLAG_PERMANENT, + CTLTYPE_NODE, "raw", + SYSCTL_DESCR("Raw IPv4 settings"), + NULL, 0, NULL, 0, + CTL_NET, PF_INET, IPPROTO_RAW, CTL_EOL); + + sysctl_createv(clog, 0, NULL, NULL, + CTLFLAG_PERMANENT, + CTLTYPE_STRUCT, "pcblist", + SYSCTL_DESCR("Raw IPv4 control block list"), + sysctl_inpcblist, 0, &rawcbtable, 0, + CTL_NET, PF_INET, IPPROTO_RAW, + CTL_CREATE, CTL_EOL); +}