NetBSD/sys/netinet/fil.c

2089 lines
47 KiB
C

/* $NetBSD: fil.c,v 1.43 2000/11/12 19:50:47 thorpej Exp $ */
/*
* Copyright (C) 1993-2000 by Darren Reed.
*
* Redistribution and use in source and binary forms are permitted
* provided that this notice is preserved and due credit is given
* to the original author and the contributors.
*/
#if !defined(lint)
#if defined(__NetBSD__)
static const char rcsid[] = "$NetBSD: fil.c,v 1.43 2000/11/12 19:50:47 thorpej Exp $";
#else
static const char sccsid[] = "@(#)fil.c 1.36 6/5/96 (C) 1993-2000 Darren Reed";
static const char rcsid[] = "@(#)Id: fil.c,v 2.35.2.19 2000/07/27 13:08:18 darrenr Exp";
#endif
#endif
#include <sys/errno.h>
#include <sys/types.h>
#include <sys/param.h>
#include <sys/time.h>
#include <sys/file.h>
#if defined(__NetBSD__) && (NetBSD >= 199905) && !defined(IPFILTER_LKM) && \
defined(_KERNEL) && !defined(_LKM)
# include "opt_ipfilter_log.h"
#endif
#if (defined(KERNEL) || defined(_KERNEL)) && defined(__FreeBSD_version) && \
(__FreeBSD_version >= 220000)
# if (__FreeBSD_version >= 400000)
# ifndef KLD_MODULE
# include "opt_inet6.h"
# endif
# if (__FreeBSD_version == 400019)
# define CSUM_DELAY_DATA
# endif
# endif
# include <sys/filio.h>
# include <sys/fcntl.h>
#else
# include <sys/ioctl.h>
#endif
#if (defined(_KERNEL) || defined(KERNEL)) && !defined(linux)
# include <sys/systm.h>
#else
# include <stdio.h>
# include <string.h>
# include <stdlib.h>
#endif
#include <sys/uio.h>
#if !defined(__SVR4) && !defined(__svr4__)
# ifndef linux
# include <sys/mbuf.h>
# endif
#else
# include <sys/byteorder.h>
# if SOLARIS2 < 5
# include <sys/dditypes.h>
# endif
# include <sys/stream.h>
#endif
#ifndef linux
# include <sys/protosw.h>
# include <sys/socket.h>
#endif
#include <net/if.h>
#ifdef sun
# include <net/af.h>
#endif
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#ifndef linux
# include <netinet/ip_var.h>
#endif
#if defined(__sgi) && defined(IFF_DRVRLOCK) /* IRIX 6 */
# include <sys/hashing.h>
# include <netinet/in_var.h>
#endif
#include <netinet/tcp.h>
#include <netinet/udp.h>
#include <netinet/ip_icmp.h>
#include "netinet/ip_compat.h"
#ifdef USE_INET6
# include <netinet/icmp6.h>
# if !SOLARIS && defined(_KERNEL)
# include <netinet6/in6_var.h>
# endif
#endif
#include <netinet/tcpip.h>
#include "netinet/ip_fil.h"
#include "netinet/ip_proxy.h"
#include "netinet/ip_nat.h"
#include "netinet/ip_frag.h"
#include "netinet/ip_state.h"
#include "netinet/ip_auth.h"
# if defined(__FreeBSD_version) && (__FreeBSD_version >= 300000)
# include <sys/malloc.h>
# if defined(_KERNEL) && !defined(IPFILTER_LKM)
# include "opt_ipfilter.h"
# endif
# endif
#ifndef MIN
# define MIN(a,b) (((a)<(b))?(a):(b))
#endif
#include "netinet/ipl.h"
#ifndef _KERNEL
# include "ipf.h"
# include "ipt.h"
extern int opts;
# define FR_VERBOSE(verb_pr) verbose verb_pr
# define FR_DEBUG(verb_pr) debug verb_pr
# define IPLLOG(a, c, d, e) ipllog()
#else /* #ifndef _KERNEL */
# define FR_VERBOSE(verb_pr)
# define FR_DEBUG(verb_pr)
# define IPLLOG(a, c, d, e) ipflog(a, c, d, e)
# if SOLARIS || defined(__sgi)
extern KRWLOCK_T ipf_mutex, ipf_auth, ipf_nat;
extern kmutex_t ipf_rw;
# endif
# if SOLARIS
# define FR_NEWAUTH(m, fi, ip, qif) fr_newauth((mb_t *)m, fi, \
ip, qif)
# else /* SOLARIS */
# define FR_NEWAUTH(m, fi, ip, qif) fr_newauth((mb_t *)m, fi, ip)
# endif /* SOLARIS || __sgi */
#endif /* _KERNEL */
struct filterstats frstats[2] = {{0,0,0,0,0},{0,0,0,0,0}};
struct frentry *ipfilter[2][2] = { { NULL, NULL }, { NULL, NULL } },
#ifdef USE_INET6
*ipfilter6[2][2] = { { NULL, NULL }, { NULL, NULL } },
*ipacct6[2][2] = { { NULL, NULL }, { NULL, NULL } },
#endif
*ipacct[2][2] = { { NULL, NULL }, { NULL, NULL } };
struct frgroup *ipfgroups[3][2];
int fr_flags = IPF_LOGGING;
int fr_active = 0;
int fr_chksrc = 0;
#if defined(IPFILTER_DEFAULT_BLOCK)
int fr_pass = FR_NOMATCH|FR_BLOCK;
#else
int fr_pass = (IPF_DEFAULT_PASS|FR_NOMATCH);
#endif
char ipfilter_version[] = IPL_VERSION;
fr_info_t frcache[2];
static int frflushlist __P((int, minor_t, int *, frentry_t **));
#ifdef _KERNEL
static void frsynclist __P((frentry_t *));
#endif
/*
* bit values for identifying presence of individual IP options
*/
struct optlist ipopts[20] = {
{ IPOPT_NOP, 0x000001 },
{ IPOPT_RR, 0x000002 },
{ IPOPT_ZSU, 0x000004 },
{ IPOPT_MTUP, 0x000008 },
{ IPOPT_MTUR, 0x000010 },
{ IPOPT_ENCODE, 0x000020 },
{ IPOPT_TS, 0x000040 },
{ IPOPT_TR, 0x000080 },
{ IPOPT_SECURITY, 0x000100 },
{ IPOPT_LSRR, 0x000200 },
{ IPOPT_E_SEC, 0x000400 },
{ IPOPT_CIPSO, 0x000800 },
{ IPOPT_SATID, 0x001000 },
{ IPOPT_SSRR, 0x002000 },
{ IPOPT_ADDEXT, 0x004000 },
{ IPOPT_VISA, 0x008000 },
{ IPOPT_IMITD, 0x010000 },
{ IPOPT_EIP, 0x020000 },
{ IPOPT_FINN, 0x040000 },
{ 0, 0x000000 }
};
/*
* bit values for identifying presence of individual IP security options
*/
struct optlist secopt[8] = {
{ IPSO_CLASS_RES4, 0x01 },
{ IPSO_CLASS_TOPS, 0x02 },
{ IPSO_CLASS_SECR, 0x04 },
{ IPSO_CLASS_RES3, 0x08 },
{ IPSO_CLASS_CONF, 0x10 },
{ IPSO_CLASS_UNCL, 0x20 },
{ IPSO_CLASS_RES2, 0x40 },
{ IPSO_CLASS_RES1, 0x80 }
};
/*
* compact the IP header into a structure which contains just the info.
* which is useful for comparing IP headers with.
*/
void fr_makefrip(hlen, ip, fin)
int hlen;
ip_t *ip;
fr_info_t *fin;
{
u_short optmsk = 0, secmsk = 0, auth = 0;
int i, mv, ol, off, p, plen, v;
fr_ip_t *fi = &fin->fin_fi;
struct optlist *op;
u_char *s, opt;
tcphdr_t *tcp;
fin->fin_rev = 0;
fin->fin_fr = NULL;
fin->fin_tcpf = 0;
fin->fin_data[0] = 0;
fin->fin_data[1] = 0;
fin->fin_rule = -1;
fin->fin_group = -1;
#ifdef _KERNEL
fin->fin_icode = ipl_unreach;
#endif
v = fin->fin_v;
fi->fi_v = v;
fin->fin_hlen = hlen;
if (v == 4) {
fin->fin_id = ip->ip_id;
fi->fi_tos = ip->ip_tos;
off = (ip->ip_off & IP_OFFMASK) << 3;
tcp = (tcphdr_t *)((char *)ip + hlen);
(*(((u_short *)fi) + 1)) = (*(((u_short *)ip) + 4));
fi->fi_src.i6[1] = 0;
fi->fi_src.i6[2] = 0;
fi->fi_src.i6[3] = 0;
fi->fi_dst.i6[1] = 0;
fi->fi_dst.i6[2] = 0;
fi->fi_dst.i6[3] = 0;
fi->fi_saddr = ip->ip_src.s_addr;
fi->fi_daddr = ip->ip_dst.s_addr;
p = ip->ip_p;
fi->fi_fl = (hlen > sizeof(ip_t)) ? FI_OPTIONS : 0;
if (ip->ip_off & 0x3fff)
fi->fi_fl |= FI_FRAG;
plen = ip->ip_len;
fin->fin_dlen = plen - hlen;
}
#ifdef USE_INET6
else if (v == 6) {
ip6_t *ip6 = (ip6_t *)ip;
off = 0;
p = ip6->ip6_nxt;
fi->fi_p = p;
fi->fi_ttl = ip6->ip6_hlim;
tcp = (tcphdr_t *)(ip6 + 1);
fi->fi_src.in6 = ip6->ip6_src;
fi->fi_dst.in6 = ip6->ip6_dst;
fin->fin_id = (u_short)(ip6->ip6_flow & 0xffff);
fi->fi_tos = 0;
fi->fi_fl = 0;
plen = ntohs(ip6->ip6_plen);
fin->fin_dlen = plen;
}
#endif
else
return;
fin->fin_off = off;
fin->fin_plen = plen;
fin->fin_dp = (void *)tcp;
switch (p)
{
case IPPROTO_ICMP :
{
int minicmpsz = sizeof(struct icmp);
icmphdr_t *icmp;
icmp = (icmphdr_t *)tcp;
if (!off && (icmp->icmp_type == ICMP_ECHOREPLY ||
icmp->icmp_type == ICMP_ECHO))
minicmpsz = ICMP_MINLEN;
/* type(1) + code(1) + cksum(2) + id(2) seq(2) +
* 3*timestamp(3*4) */
else if (!off && (icmp->icmp_type == ICMP_TSTAMP ||
icmp->icmp_type == ICMP_TSTAMPREPLY))
minicmpsz = 20;
/* type(1) + code(1) + cksum(2) + id(2) seq(2) + mask(4) */
else if (!off && (icmp->icmp_type == ICMP_MASKREQ ||
icmp->icmp_type == ICMP_MASKREPLY))
minicmpsz = 12;
if ((!(plen >= hlen + minicmpsz) && !off) ||
(off && off < sizeof(struct icmp)))
fi->fi_fl |= FI_SHORT;
if (fin->fin_dlen > 1)
fin->fin_data[0] = *(u_short *)tcp;
break;
}
case IPPROTO_TCP :
fi->fi_fl |= FI_TCPUDP;
#ifdef USE_INET6
if (v == 6) {
if (plen < sizeof(struct tcphdr))
fi->fi_fl |= FI_SHORT;
} else
#endif
if (v == 4) {
if ((!IPMINLEN(ip, tcphdr) && !off) ||
(off && off < sizeof(struct tcphdr)))
fi->fi_fl |= FI_SHORT;
}
if (!(fi->fi_fl & FI_SHORT) && !off)
fin->fin_tcpf = tcp->th_flags;
goto getports;
case IPPROTO_UDP :
fi->fi_fl |= FI_TCPUDP;
#ifdef USE_INET6
if (v == 6) {
if (plen < sizeof(struct udphdr))
fi->fi_fl |= FI_SHORT;
} else
#endif
if (v == 4) {
if ((!IPMINLEN(ip, udphdr) && !off) ||
(off && off < sizeof(struct udphdr)))
fi->fi_fl |= FI_SHORT;
}
getports:
if (!off && (fin->fin_dlen > 3)) {
fin->fin_data[0] = ntohs(tcp->th_sport);
fin->fin_data[1] = ntohs(tcp->th_dport);
}
break;
default :
break;
}
#ifdef USE_INET6
if (v == 6) {
fi->fi_optmsk = 0;
fi->fi_secmsk = 0;
fi->fi_auth = 0;
return;
}
#endif
for (s = (u_char *)(ip + 1), hlen -= (int)sizeof(*ip); hlen > 0; ) {
opt = *s;
if (opt == '\0')
break;
else if (opt == IPOPT_NOP)
ol = 1;
else {
if (hlen < 2)
break;
ol = (int)*(s + 1);
if (ol < 2 || ol > hlen)
break;
}
for (i = 9, mv = 4; mv >= 0; ) {
op = ipopts + i;
if (opt == (u_char)op->ol_val) {
optmsk |= op->ol_bit;
if (opt == IPOPT_SECURITY) {
struct optlist *sp;
u_char sec;
int j, m;
sec = *(s + 2); /* classification */
for (j = 3, m = 2; m >= 0; ) {
sp = secopt + j;
if (sec == sp->ol_val) {
secmsk |= sp->ol_bit;
auth = *(s + 3);
auth *= 256;
auth += *(s + 4);
break;
}
if (sec < sp->ol_val)
j -= m--;
else
j += m--;
}
}
break;
}
if (opt < op->ol_val)
i -= mv--;
else
i += mv--;
}
hlen -= ol;
s += ol;
}
if (auth && !(auth & 0x0100))
auth &= 0xff00;
fi->fi_optmsk = optmsk;
fi->fi_secmsk = secmsk;
fi->fi_auth = auth;
}
/*
* check an IP packet for TCP/UDP characteristics such as ports and flags.
*/
int fr_tcpudpchk(ft, fin)
frtuc_t *ft;
fr_info_t *fin;
{
u_short po, tup;
char i;
int err = 1;
/*
* Both ports should *always* be in the first fragment.
* So far, I cannot find any cases where they can not be.
*
* compare destination ports
*/
if ((i = (int)ft->ftu_dcmp)) {
po = ft->ftu_dport;
tup = fin->fin_data[1];
/*
* Do opposite test to that required and
* continue if that succeeds.
*/
if (!--i && tup != po) /* EQUAL */
err = 0;
else if (!--i && tup == po) /* NOTEQUAL */
err = 0;
else if (!--i && tup >= po) /* LESSTHAN */
err = 0;
else if (!--i && tup <= po) /* GREATERTHAN */
err = 0;
else if (!--i && tup > po) /* LT or EQ */
err = 0;
else if (!--i && tup < po) /* GT or EQ */
err = 0;
else if (!--i && /* Out of range */
(tup >= po && tup <= ft->ftu_dtop))
err = 0;
else if (!--i && /* In range */
(tup <= po || tup >= ft->ftu_dtop))
err = 0;
}
/*
* compare source ports
*/
if (err && (i = (int)ft->ftu_scmp)) {
po = ft->ftu_sport;
tup = fin->fin_data[0];
if (!--i && tup != po)
err = 0;
else if (!--i && tup == po)
err = 0;
else if (!--i && tup >= po)
err = 0;
else if (!--i && tup <= po)
err = 0;
else if (!--i && tup > po)
err = 0;
else if (!--i && tup < po)
err = 0;
else if (!--i && /* Out of range */
(tup >= po && tup <= ft->ftu_stop))
err = 0;
else if (!--i && /* In range */
(tup <= po || tup >= ft->ftu_stop))
err = 0;
}
/*
* If we don't have all the TCP/UDP header, then how can we
* expect to do any sort of match on it ? If we were looking for
* TCP flags, then NO match. If not, then match (which should
* satisfy the "short" class too).
*/
if (err && (fin->fin_fi.fi_p == IPPROTO_TCP)) {
if (fin->fin_fi.fi_fl & FI_SHORT)
return !(ft->ftu_tcpf | ft->ftu_tcpfm);
/*
* Match the flags ? If not, abort this match.
*/
if (ft->ftu_tcpfm &&
ft->ftu_tcpf != (fin->fin_tcpf & ft->ftu_tcpfm)) {
FR_DEBUG(("f. %#x & %#x != %#x\n", fin->fin_tcpf,
ft->ftu_tcpfm, ft->ftu_tcpf));
err = 0;
}
}
return err;
}
/*
* Check the input/output list of rules for a match and result.
* Could be per interface, but this gets real nasty when you don't have
* kernel sauce.
*/
int fr_scanlist(pass, ip, fin, m)
u_32_t pass;
ip_t *ip;
fr_info_t *fin;
void *m;
{
struct frentry *fr;
fr_ip_t *fi = &fin->fin_fi;
int rulen, portcmp = 0, off, skip = 0, logged = 0;
u_32_t passt;
fr = fin->fin_fr;
fin->fin_fr = NULL;
fin->fin_rule = 0;
fin->fin_group = 0;
if (fin->fin_v == 4)
off = ip->ip_off & IP_OFFMASK;
else
off = 0;
pass |= (fi->fi_fl << 24);
if ((fi->fi_fl & FI_TCPUDP) && (fin->fin_dlen > 3) && !off)
portcmp = 1;
for (rulen = 0; fr; fr = fr->fr_next, rulen++) {
if (skip) {
skip--;
continue;
}
/*
* In all checks below, a null (zero) value in the
* filter struture is taken to mean a wildcard.
*
* check that we are working for the right interface
*/
#ifdef _KERNEL
# if BSD >= 199306
if (fin->fin_out != 0) {
if ((fr->fr_oifa &&
fr->fr_oifa != ((mb_t *)m)->m_pkthdr.rcvif) ||
(fr->fr_ifa && fr->fr_ifa != fin->fin_ifp))
continue;
} else
# endif
if (fr->fr_ifa && fr->fr_ifa != fin->fin_ifp)
continue;
#else
if (opts & (OPT_VERBOSE|OPT_DEBUG))
printf("\n");
FR_VERBOSE(("%c", (pass & FR_PASS) ? 'p' :
(pass & FR_AUTH) ? 'a' : 'b'));
if (fr->fr_ifa && fr->fr_ifa != fin->fin_ifp)
continue;
FR_VERBOSE((":i"));
#endif
{
u_32_t *ld, *lm, *lip;
int i;
lip = (u_32_t *)fi;
lm = (u_32_t *)&fr->fr_mip;
ld = (u_32_t *)&fr->fr_ip;
i = ((*lip & *lm) != *ld);
FR_DEBUG(("0. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
if (i)
continue;
/*
* We now know whether the packet version and the
* rule version match, along with protocol, ttl and
* tos.
*/
lip++, lm++, ld++;
/*
* Unrolled loops (4 each, for 32 bits).
*/
i |= ((*lip & *lm) != *ld) << 19;
FR_DEBUG(("1a. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
if (fi->fi_v == 6) {
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 19;
FR_DEBUG(("1b. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 19;
FR_DEBUG(("1c. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 19;
FR_DEBUG(("1d. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
} else {
lip += 3;
lm += 3;
ld += 3;
}
i ^= (fr->fr_flags & FR_NOTSRCIP);
if (i)
continue;
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 20;
FR_DEBUG(("2a. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
if (fi->fi_v == 6) {
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 20;
FR_DEBUG(("2b. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 20;
FR_DEBUG(("2c. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld) << 20;
FR_DEBUG(("2d. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
} else {
lip += 3;
lm += 3;
ld += 3;
}
i ^= (fr->fr_flags & FR_NOTDSTIP);
if (i)
continue;
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld);
FR_DEBUG(("3. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
lip++, lm++, ld++;
i |= ((*lip & *lm) != *ld);
FR_DEBUG(("4. %#08x & %#08x != %#08x\n",
*lip, *lm, *ld));
if (i)
continue;
}
/*
* If a fragment, then only the first has what we're looking
* for here...
*/
if (!portcmp && (fr->fr_dcmp || fr->fr_scmp || fr->fr_tcpf ||
fr->fr_tcpfm))
continue;
if (fi->fi_fl & FI_TCPUDP) {
if (!fr_tcpudpchk(&fr->fr_tuc, fin))
continue;
} else if (fr->fr_icmpm || fr->fr_icmp) {
if ((fi->fi_p != IPPROTO_ICMP) || off ||
(fin->fin_dlen < 2))
continue;
if ((fin->fin_data[0] & fr->fr_icmpm) != fr->fr_icmp) {
FR_DEBUG(("i. %#x & %#x != %#x\n",
fin->fin_data[0], fr->fr_icmpm,
fr->fr_icmp));
continue;
}
}
FR_VERBOSE(("*"));
/*
* Just log this packet...
*/
passt = fr->fr_flags;
#if (BSD >= 199306) && (defined(_KERNEL) || defined(KERNEL))
if (securelevel <= 0)
#endif
if ((passt & FR_CALLNOW) && fr->fr_func)
passt = (*fr->fr_func)(passt, ip, fin);
fin->fin_fr = fr;
#ifdef IPFILTER_LOG
if ((passt & FR_LOGMASK) == FR_LOG) {
if (!IPLLOG(passt, ip, fin, m)) {
if (passt & FR_LOGORBLOCK)
passt |= FR_BLOCK|FR_QUICK;
ATOMIC_INCL(frstats[fin->fin_out].fr_skip);
}
ATOMIC_INCL(frstats[fin->fin_out].fr_pkl);
logged = 1;
}
#endif /* IPFILTER_LOG */
if (!(skip = fr->fr_skip) && (passt & FR_LOGMASK) != FR_LOG)
pass = passt;
FR_DEBUG(("pass %#x\n", pass));
ATOMIC_INCL(fr->fr_hits);
if (pass & FR_ACCOUNT)
fr->fr_bytes += (U_QUAD_T)ip->ip_len;
else
fin->fin_icode = fr->fr_icode;
fin->fin_rule = rulen;
fin->fin_group = fr->fr_group;
if (fr->fr_grp) {
fin->fin_fr = fr->fr_grp;
pass = fr_scanlist(pass, ip, fin, m);
if (fin->fin_fr == NULL) {
fin->fin_rule = rulen;
fin->fin_group = fr->fr_group;
fin->fin_fr = fr;
}
if (pass & FR_DONTCACHE)
logged = 1;
}
if (pass & FR_QUICK)
break;
}
if (logged)
pass |= FR_DONTCACHE;
return pass;
}
#if defined(__NetBSD_Version__) && (__NetBSD_Version__ >= 105110000) && \
defined(_KERNEL)
#include <net/pfil.h>
int
fr_check_wrapper(void *arg, struct mbuf **mp, struct ifnet *ifp, int dir)
{
struct ip *ip = mtod(*mp, struct ip *);
int rv, hlen = ip->ip_hl << 2;
/*
* We get the packet with all fields in network byte
* order. We expect ip_len and ip_off to be in host
* order. We frob them, call the filter, then frob
* them back.
*
* Note, we don't need to update the checksum, because
* it has already been verified.
*/
NTOHS(ip->ip_len);
NTOHS(ip->ip_off);
rv = fr_check(ip, hlen, ifp, (dir == PFIL_OUT), mp);
if (rv == 0 && *mp != NULL) {
ip = mtod(*mp, struct ip *);
HTONS(ip->ip_len);
HTONS(ip->ip_off);
}
return (rv);
}
#ifdef USE_INET6
#include <netinet/ip6.h>
int
fr_check_wrapper6(void *arg, struct mbuf **mp, struct ifnet *ifp, int dir)
{
return (fr_check(mtod(*mp, struct ip *), sizeof(struct ip6_hdr),
ifp, (dir == PFIL_OUT), mp));
}
#endif
#endif /* __NetBSD_Version >= 105110000 && _KERNEL */
/*
* frcheck - filter check
* check using source and destination addresses/ports in a packet whether
* or not to pass it on or not.
*/
int fr_check(ip, hlen, ifp, out
#if defined(_KERNEL) && SOLARIS
, qif, mp)
qif_t *qif;
#else
, mp)
#endif
mb_t **mp;
ip_t *ip;
int hlen;
void *ifp;
int out;
{
/*
* The above really sucks, but short of writing a diff
*/
fr_info_t frinfo, *fc;
register fr_info_t *fin = &frinfo;
int changed, error = EHOSTUNREACH, v = ip->ip_v;
frentry_t *fr = NULL, *list;
u_32_t pass, apass;
#if !SOLARIS || !defined(_KERNEL)
mb_t *m = *mp;
#endif
#ifdef _KERNEL
mb_t *mc = NULL;
# if !defined(__SVR4) && !defined(__svr4__)
# ifdef __sgi
char hbuf[(0xf << 2) + sizeof(struct icmp) + sizeof(ip_t) + 8];
# endif
int up;
# ifdef M_CANFASTFWD
/*
* XXX For now, IP Filter and fast-forwarding of cached flows
* XXX are mutually exclusive. Eventually, IP Filter should
* XXX get a "can-fast-forward" filter rule.
*/
m->m_flags &= ~M_CANFASTFWD;
# endif /* M_CANFASTFWD */
# ifdef CSUM_DELAY_DATA
/*
* disable delayed checksums.
*/
if (m->m_pkthdr.csum_flags & CSUM_DELAY_DATA) {
in_delayed_cksum(m);
m->m_pkthdr.csum_flags &= ~CSUM_DELAY_DATA;
}
# endif /* CSUM_DELAY_DATA */
/* at this moment, ipfilter supports IPv4 traffic only. */
if (ip->ip_v != 4)
return 0;
if ((ip->ip_p == IPPROTO_TCP || ip->ip_p == IPPROTO_UDP ||
ip->ip_p == IPPROTO_ICMP)) {
int plen = 0;
if ((ip->ip_off & IP_OFFMASK) == 0)
switch(ip->ip_p)
{
case IPPROTO_TCP:
plen = sizeof(tcphdr_t);
break;
case IPPROTO_UDP:
plen = sizeof(udphdr_t);
break;
/* 96 - enough for complete ICMP error IP header */
case IPPROTO_ICMP:
plen = ICMPERR_MAXPKTLEN - sizeof(ip_t);
break;
}
up = MIN(hlen + plen, ip->ip_len);
if (up > m->m_len) {
# ifdef __sgi
/* Under IRIX, avoid m_pullup as it makes ping <hostname> panic */
if ((up > sizeof(hbuf)) || (m_length(m) < up)) {
ATOMIC_INCL(frstats[out].fr_pull[1]);
return -1;
}
m_copydata(m, 0, up, hbuf);
ATOMIC_INCL(frstats[out].fr_pull[0]);
ip = (ip_t *)hbuf;
# else /* __ sgi */
# ifndef linux
if ((*mp = m_pullup(m, up)) == 0) {
ATOMIC_INCL(frstats[out].fr_pull[1]);
return -1;
} else {
ATOMIC_INCL(frstats[out].fr_pull[0]);
m = *mp;
ip = mtod(m, ip_t *);
}
# endif /* !linux */
# endif /* __sgi */
} else
up = 0;
} else
up = 0;
# endif /* !defined(__SVR4) && !defined(__svr4__) */
# if SOLARIS
mb_t *m = qif->qf_m;
if ((u_int)ip & 0x3)
return 2;
fin->fin_qfm = m;
fin->fin_qif = qif;
# endif
#endif /* _KERNEL */
/*
* Be careful here: ip_id is in network byte order when called
* from ip_output()
*/
if ((out) && (v == 4))
ip->ip_id = ntohs(ip->ip_id);
changed = 0;
fin->fin_v = v;
fin->fin_ifp = ifp;
fin->fin_out = out;
fin->fin_mp = mp;
fr_makefrip(hlen, ip, fin);
#ifdef _KERNEL
# ifdef USE_INET6
if (v == 6) {
ATOMIC_INCL(frstats[0].fr_ipv6[out]);
} else
# endif
# ifdef IPFILTER_LOG
if (!out && fr_chksrc && !fr_verifysrc(ip->ip_src, ifp)) {
ATOMIC_INCL(frstats[0].fr_badsrc);
if (fr_chksrc == 2) {
fin->fin_group = -2;
pass = FR_INQUE|FR_NOMATCH|FR_LOGB;
(void) IPLLOG(pass, ip, fin, m);
}
# if !SOLARIS
m_freem(m);
# endif
return error;
}
# endif
#endif
pass = fr_pass;
if (fin->fin_fi.fi_fl & FI_SHORT) {
ATOMIC_INCL(frstats[out].fr_short);
}
READ_ENTER(&ipf_mutex);
if (fin->fin_fi.fi_fl & FI_SHORT)
ATOMIC_INCL(frstats[out].fr_short);
/*
* Check auth now. This, combined with the check below to see if apass
* is 0 is to ensure that we don't count the packet twice, which can
* otherwise occur when we reprocess it. As it is, we only count it
* after it has no auth. table matchup. This also stops NAT from
* occuring until after the packet has been auth'd.
*/
apass = fr_checkauth(ip, fin);
if (!out) {
#ifdef USE_INET6
if (v == 6)
list = ipacct6[0][fr_active];
else
#endif
list = ipacct[0][fr_active];
changed = ip_natin(ip, fin);
if (!apass && (fin->fin_fr = list) &&
(fr_scanlist(FR_NOMATCH, ip, fin, m) & FR_ACCOUNT)) {
ATOMIC_INCL(frstats[0].fr_acct);
}
}
if (apass || (!(fr = ipfr_knownfrag(ip, fin)) &&
!(fr = fr_checkstate(ip, fin)))) {
/*
* If a packet is found in the auth table, then skip checking
* the access lists for permission but we do need to consider
* the result as if it were from the ACL's.
*/
if (!apass) {
fc = frcache + out;
if (!bcmp((char *)fin, (char *)fc, FI_CSIZE)) {
/*
* copy cached data so we can unlock the mutex
* earlier.
*/
bcopy((char *)fc, (char *)fin, FI_COPYSIZE);
ATOMIC_INCL(frstats[out].fr_chit);
if ((fr = fin->fin_fr)) {
ATOMIC_INCL(fr->fr_hits);
pass = fr->fr_flags;
}
} else {
#ifdef USE_INET6
if (v == 6)
list = ipfilter6[out][fr_active];
else
#endif
list = ipfilter[out][fr_active];
if ((fin->fin_fr = list))
pass = fr_scanlist(fr_pass, ip, fin, m);
if (!(pass & (FR_KEEPSTATE|FR_DONTCACHE)))
bcopy((char *)fin, (char *)fc,
FI_COPYSIZE);
if (pass & FR_NOMATCH) {
ATOMIC_INCL(frstats[out].fr_nom);
}
}
fr = fin->fin_fr;
} else
pass = apass;
/*
* If we fail to add a packet to the authorization queue,
* then we drop the packet later. However, if it was added
* then pretend we've dropped it already.
*/
if ((pass & FR_AUTH))
if (fr_newauth((mb_t *)m, fin, ip) != 0)
#ifdef _KERNEL
m = *mp = NULL;
#else
;
#endif
if (pass & FR_PREAUTH) {
READ_ENTER(&ipf_auth);
if ((fin->fin_fr = ipauth) &&
(pass = fr_scanlist(0, ip, fin, m))) {
ATOMIC_INCL(fr_authstats.fas_hits);
} else {
ATOMIC_INCL(fr_authstats.fas_miss);
}
RWLOCK_EXIT(&ipf_auth);
}
fin->fin_fr = fr;
if ((pass & (FR_KEEPFRAG|FR_KEEPSTATE)) == FR_KEEPFRAG) {
if (fin->fin_fi.fi_fl & FI_FRAG) {
if (ipfr_newfrag(ip, fin, pass) == -1) {
ATOMIC_INCL(frstats[out].fr_bnfr);
} else {
ATOMIC_INCL(frstats[out].fr_nfr);
}
} else {
ATOMIC_INCL(frstats[out].fr_cfr);
}
}
if (pass & FR_KEEPSTATE) {
if (fr_addstate(ip, fin, 0) == NULL) {
ATOMIC_INCL(frstats[out].fr_bads);
} else {
ATOMIC_INCL(frstats[out].fr_ads);
}
}
} else if (fr != NULL) {
pass = fr->fr_flags;
if (pass & FR_LOGFIRST)
pass &= ~(FR_LOGFIRST|FR_LOG);
}
#if (BSD >= 199306) && (defined(_KERNEL) || defined(KERNEL))
if (securelevel <= 0)
#endif
if (fr && fr->fr_func && !(pass & FR_CALLNOW))
pass = (*fr->fr_func)(pass, ip, fin);
/*
* Only count/translate packets which will be passed on, out the
* interface.
*/
if (out && (pass & FR_PASS)) {
#ifdef USE_INET6
if (v == 6)
list = ipacct6[1][fr_active];
else
#endif
list = ipacct[1][fr_active];
if ((fin->fin_fr = list) &&
(fr_scanlist(FR_NOMATCH, ip, fin, m) & FR_ACCOUNT)) {
ATOMIC_INCL(frstats[1].fr_acct);
}
fin->fin_fr = fr;
changed = ip_natout(ip, fin);
} else
fin->fin_fr = fr;
RWLOCK_EXIT(&ipf_mutex);
#ifdef IPFILTER_LOG
if ((fr_flags & FF_LOGGING) || (pass & FR_LOGMASK)) {
if ((fr_flags & FF_LOGNOMATCH) && (pass & FR_NOMATCH)) {
pass |= FF_LOGNOMATCH;
ATOMIC_INCL(frstats[out].fr_npkl);
goto logit;
} else if (((pass & FR_LOGMASK) == FR_LOGP) ||
((pass & FR_PASS) && (fr_flags & FF_LOGPASS))) {
if ((pass & FR_LOGMASK) != FR_LOGP)
pass |= FF_LOGPASS;
ATOMIC_INCL(frstats[out].fr_ppkl);
goto logit;
} else if (((pass & FR_LOGMASK) == FR_LOGB) ||
((pass & FR_BLOCK) && (fr_flags & FF_LOGBLOCK))) {
if ((pass & FR_LOGMASK) != FR_LOGB)
pass |= FF_LOGBLOCK;
ATOMIC_INCL(frstats[out].fr_bpkl);
logit:
if (!IPLLOG(pass, ip, fin, m)) {
ATOMIC_INCL(frstats[out].fr_skip);
if ((pass & (FR_PASS|FR_LOGORBLOCK)) ==
(FR_PASS|FR_LOGORBLOCK))
pass ^= FR_PASS|FR_BLOCK;
}
}
}
#endif /* IPFILTER_LOG */
if ((out) && (v == 4))
ip->ip_id = htons(ip->ip_id);
#ifdef _KERNEL
/*
* Only allow FR_DUP to work if a rule matched - it makes no sense to
* set FR_DUP as a "default" as there are no instructions about where
* to send the packet.
*/
if (fr && (pass & FR_DUP))
# if SOLARIS
mc = dupmsg(m);
# else
# ifndef linux
mc = m_copy(m, 0, M_COPYALL);
# else
;
# endif
# endif
#endif
if (pass & FR_PASS) {
ATOMIC_INCL(frstats[out].fr_pass);
} else if (pass & FR_BLOCK) {
ATOMIC_INCL(frstats[out].fr_block);
/*
* Should we return an ICMP packet to indicate error
* status passing through the packet filter ?
* WARNING: ICMP error packets AND TCP RST packets should
* ONLY be sent in repsonse to incoming packets. Sending them
* in response to outbound packets can result in a panic on
* some operating systems.
*/
if (!out) {
#ifdef _KERNEL
if (pass & FR_RETICMP) {
int dst;
if ((pass & FR_RETMASK) == FR_FAKEICMP)
dst = 1;
else
dst = 0;
send_icmp_err(ip, ICMP_UNREACH, fin, dst);
ATOMIC_INCL(frstats[0].fr_ret);
} else if (((pass & FR_RETMASK) == FR_RETRST) &&
!(fin->fin_fi.fi_fl & FI_SHORT)) {
if (send_reset(ip, fin) == 0) {
ATOMIC_INCL(frstats[1].fr_ret);
}
}
#else
if ((pass & FR_RETMASK) == FR_RETICMP) {
verbose("- ICMP unreachable sent\n");
ATOMIC_INCL(frstats[0].fr_ret);
} else if ((pass & FR_RETMASK) == FR_FAKEICMP) {
verbose("- forged ICMP unreachable sent\n");
ATOMIC_INCL(frstats[0].fr_ret);
} else if (((pass & FR_RETMASK) == FR_RETRST) &&
!(fin->fin_fi.fi_fl & FI_SHORT)) {
verbose("- TCP RST sent\n");
ATOMIC_INCL(frstats[1].fr_ret);
}
#endif
} else {
if (pass & FR_RETRST)
error = ECONNRESET;
}
}
/*
* If we didn't drop off the bottom of the list of rules (and thus
* the 'current' rule fr is not NULL), then we may have some extra
* instructions about what to do with a packet.
* Once we're finished return to our caller, freeing the packet if
* we are dropping it (* BSD ONLY *).
*/
if ((changed == -1) && (pass & FR_PASS)) {
pass &= ~FR_PASS;
pass |= FR_BLOCK;
}
#if defined(_KERNEL)
# if !SOLARIS
# if !defined(linux)
if (fr) {
frdest_t *fdp = &fr->fr_tif;
if (((pass & FR_FASTROUTE) && !out) ||
(fdp->fd_ifp && fdp->fd_ifp != (struct ifnet *)-1)) {
if (ipfr_fastroute(m, fin, fdp) == 0)
m = *mp = NULL;
}
if (mc)
ipfr_fastroute(mc, fin, &fr->fr_dif);
}
if (!(pass & FR_PASS) && m)
m_freem(m);
# ifdef __sgi
else if (changed && up && m)
m_copyback(m, 0, up, hbuf);
# endif
# endif /* !linux */
# else /* !SOLARIS */
if (fr) {
frdest_t *fdp = &fr->fr_tif;
if (((pass & FR_FASTROUTE) && !out) ||
(fdp->fd_ifp && fdp->fd_ifp != (struct ifnet *)-1)) {
if (ipfr_fastroute(ip, m, mp, fin, fdp) == 0)
m = *mp = NULL;
}
if (mc)
ipfr_fastroute(ip, mc, mp, fin, &fr->fr_dif);
}
# endif /* !SOLARIS */
return (pass & FR_PASS) ? 0 : error;
#else /* _KERNEL */
if (pass & FR_NOMATCH)
return 1;
if (pass & FR_PASS)
return 0;
if (pass & FR_AUTH)
return -2;
return -1;
#endif /* _KERNEL */
}
/*
* ipf_cksum
* addr should be 16bit aligned and len is in bytes.
* length is in bytes
*/
u_short ipf_cksum(addr, len)
u_short *addr;
int len;
{
u_32_t sum = 0;
for (sum = 0; len > 1; len -= 2)
sum += *addr++;
/* mop up an odd byte, if necessary */
if (len == 1)
sum += *(u_char *)addr;
/*
* add back carry outs from top 16 bits to low 16 bits
*/
sum = (sum >> 16) + (sum & 0xffff); /* add hi 16 to low 16 */
sum += (sum >> 16); /* add carry */
return (u_short)(~sum);
}
/*
* NB: This function assumes we've pullup'd enough for all of the IP header
* and the TCP header. We also assume that data blocks aren't allocated in
* odd sizes.
*/
u_short fr_tcpsum(m, ip, tcp)
mb_t *m;
ip_t *ip;
tcphdr_t *tcp;
{
u_short *sp, slen, ts;
u_int sum, sum2;
int hlen;
/*
* Add up IP Header portion
*/
hlen = ip->ip_hl << 2;
slen = ip->ip_len - hlen;
sum = htons((u_short)ip->ip_p);
sum += htons(slen);
sp = (u_short *)&ip->ip_src;
sum += *sp++; /* ip_src */
sum += *sp++;
sum += *sp++; /* ip_dst */
sum += *sp++;
ts = tcp->th_sum;
tcp->th_sum = 0;
#ifdef KERNEL
# if SOLARIS
sum2 = ip_cksum(m, hlen, sum); /* hlen == offset */
sum2 = (sum2 & 0xffff) + (sum2 >> 16);
sum2 = ~sum2 & 0xffff;
# else /* SOLARIS */
# if defined(BSD) || defined(sun)
# if BSD >= 199306
m->m_data += hlen;
# else
m->m_off += hlen;
# endif
m->m_len -= hlen;
sum2 = in_cksum(m, slen);
m->m_len += hlen;
# if BSD >= 199306
m->m_data -= hlen;
# else
m->m_off -= hlen;
# endif
/*
* Both sum and sum2 are partial sums, so combine them together.
*/
sum = (sum & 0xffff) + (sum >> 16);
sum = ~sum & 0xffff;
sum2 += sum;
sum2 = (sum2 & 0xffff) + (sum2 >> 16);
# else /* defined(BSD) || defined(sun) */
{
union {
u_char c[2];
u_short s;
} bytes;
u_short len = ip->ip_len;
# if defined(__sgi)
int add;
# endif
/*
* Add up IP Header portion
*/
sp = (u_short *)&ip->ip_src;
len -= (ip->ip_hl << 2);
sum = ntohs(IPPROTO_TCP);
sum += htons(len);
sum += *sp++; /* ip_src */
sum += *sp++;
sum += *sp++; /* ip_dst */
sum += *sp++;
if (sp != (u_short *)tcp)
sp = (u_short *)tcp;
sum += *sp++; /* sport */
sum += *sp++; /* dport */
sum += *sp++; /* seq */
sum += *sp++;
sum += *sp++; /* ack */
sum += *sp++;
sum += *sp++; /* off */
sum += *sp++; /* win */
sum += *sp++; /* Skip over checksum */
sum += *sp++; /* urp */
# ifdef __sgi
/*
* In case we had to copy the IP & TCP header out of mbufs,
* skip over the mbuf bits which are the header
*/
if ((caddr_t)ip != mtod(m, caddr_t)) {
hlen = (caddr_t)sp - (caddr_t)ip;
while (hlen) {
add = MIN(hlen, m->m_len);
sp = (u_short *)(mtod(m, caddr_t) + add);
hlen -= add;
if (add == m->m_len) {
m = m->m_next;
if (!hlen) {
if (!m)
break;
sp = mtod(m, u_short *);
}
PANIC((!m),("fr_tcpsum(1): not enough data"));
}
}
}
# endif
if (!(len -= sizeof(*tcp)))
goto nodata;
while (len > 1) {
if (((caddr_t)sp - mtod(m, caddr_t)) >= m->m_len) {
m = m->m_next;
PANIC((!m),("fr_tcpsum(2): not enough data"));
sp = mtod(m, u_short *);
}
if (((caddr_t)(sp + 1) - mtod(m, caddr_t)) > m->m_len) {
bytes.c[0] = *(u_char *)sp;
m = m->m_next;
PANIC((!m),("fr_tcpsum(3): not enough data"));
sp = mtod(m, u_short *);
bytes.c[1] = *(u_char *)sp;
sum += bytes.s;
sp = (u_short *)((u_char *)sp + 1);
}
if ((u_long)sp & 1) {
bcopy((char *)sp++, (char *)&bytes.s, sizeof(bytes.s));
sum += bytes.s;
} else
sum += *sp++;
len -= 2;
}
if (len)
sum += ntohs(*(u_char *)sp << 8);
nodata:
while (sum > 0xffff)
sum = (sum & 0xffff) + (sum >> 16);
sum2 = (u_short)(~sum & 0xffff);
}
# endif /* defined(BSD) || defined(sun) */
# endif /* SOLARIS */
#else /* KERNEL */
sum2 = 0;
#endif /* KERNEL */
tcp->th_sum = ts;
return sum2;
}
#if defined(_KERNEL) && ( ((BSD < 199306) && !SOLARIS) || defined(__sgi) )
/*
* Copyright (c) 1982, 1986, 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.
*
* @(#)uipc_mbuf.c 8.2 (Berkeley) 1/4/94
* Id: fil.c,v 2.35.2.19 2000/07/27 13:08:18 darrenr Exp
*/
/*
* Copy data from an mbuf chain starting "off" bytes from the beginning,
* continuing for "len" bytes, into the indicated buffer.
*/
void
m_copydata(m, off, len, cp)
mb_t *m;
int off;
int len;
caddr_t cp;
{
unsigned count;
if (off < 0 || len < 0)
panic("m_copydata");
while (off > 0) {
if (m == 0)
panic("m_copydata");
if (off < m->m_len)
break;
off -= m->m_len;
m = m->m_next;
}
while (len > 0) {
if (m == 0)
panic("m_copydata");
count = MIN(m->m_len - off, len);
bcopy(mtod(m, caddr_t) + off, cp, count);
len -= count;
cp += count;
off = 0;
m = m->m_next;
}
}
# ifndef linux
/*
* 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;
int off;
int len;
caddr_t cp;
{
int mlen;
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;
n->m_len = min(MLEN, len);
m->m_next = n;
}
m = m->m_next;
}
out:
#if 0
if (((m = m0)->m_flags & M_PKTHDR) && (m->m_pkthdr.len < totlen))
m->m_pkthdr.len = totlen;
#endif
return;
}
# endif /* linux */
#endif /* (_KERNEL) && ( ((BSD < 199306) && !SOLARIS) || __sgi) */
frgroup_t *fr_findgroup(num, flags, which, set, fgpp)
u_32_t num, flags;
minor_t which;
int set;
frgroup_t ***fgpp;
{
frgroup_t *fg, **fgp;
if (which == IPL_LOGAUTH)
fgp = &ipfgroups[2][set];
else if (flags & FR_ACCOUNT)
fgp = &ipfgroups[1][set];
else if (flags & (FR_OUTQUE|FR_INQUE))
fgp = &ipfgroups[0][set];
else
return NULL;
num &= 0xffff;
while ((fg = *fgp))
if (fg->fg_num == num)
break;
else
fgp = &fg->fg_next;
if (fgpp)
*fgpp = fgp;
return fg;
}
frgroup_t *fr_addgroup(num, fp, which, set)
u_32_t num;
frentry_t *fp;
minor_t which;
int set;
{
frgroup_t *fg, **fgp;
if ((fg = fr_findgroup(num, fp->fr_flags, which, set, &fgp)))
return fg;
KMALLOC(fg, frgroup_t *);
if (fg) {
fg->fg_num = num;
fg->fg_next = *fgp;
fg->fg_head = fp;
fg->fg_start = &fp->fr_grp;
*fgp = fg;
}
return fg;
}
void fr_delgroup(num, flags, which, set)
u_32_t num, flags;
minor_t which;
int set;
{
frgroup_t *fg, **fgp;
if (!(fg = fr_findgroup(num, flags, which, set, &fgp)))
return;
*fgp = fg->fg_next;
KFREE(fg);
}
/*
* recursively flush rules from the list, descending groups as they are
* encountered. if a rule is the head of a group and it has lost all its
* group members, then also delete the group reference.
*/
static int frflushlist(set, unit, nfreedp, listp)
int set;
minor_t unit;
int *nfreedp;
frentry_t **listp;
{
int freed = 0, i;
frentry_t *fp;
while ((fp = *listp)) {
*listp = fp->fr_next;
if (fp->fr_grp) {
i = frflushlist(set, unit, nfreedp, &fp->fr_grp);
MUTEX_ENTER(&ipf_rw);
fp->fr_ref -= i;
MUTEX_EXIT(&ipf_rw);
}
ATOMIC_DEC32(fp->fr_ref);
if (fp->fr_grhead) {
fr_delgroup(fp->fr_grhead, fp->fr_flags,
unit, set);
fp->fr_grhead = 0;
}
if (fp->fr_ref == 0) {
KFREE(fp);
freed++;
} else
fp->fr_next = NULL;
}
*nfreedp += freed;
return freed;
}
int frflush(unit, flags)
minor_t unit;
int flags;
{
int flushed = 0, set;
if (unit != IPL_LOGIPF)
return 0;
WRITE_ENTER(&ipf_mutex);
bzero((char *)frcache, sizeof(frcache[0]) * 2);
set = fr_active;
if (flags & FR_INACTIVE)
set = 1 - set;
if (flags & FR_OUTQUE) {
#ifdef USE_INET6
(void) frflushlist(set, unit, &flushed, &ipfilter6[1][set]);
(void) frflushlist(set, unit, &flushed, &ipacct6[1][set]);
#endif
(void) frflushlist(set, unit, &flushed, &ipfilter[1][set]);
(void) frflushlist(set, unit, &flushed, &ipacct[1][set]);
}
if (flags & FR_INQUE) {
#ifdef USE_INET6
(void) frflushlist(set, unit, &flushed, &ipfilter6[0][set]);
(void) frflushlist(set, unit, &flushed, &ipacct6[0][set]);
#endif
(void) frflushlist(set, unit, &flushed, &ipfilter[0][set]);
(void) frflushlist(set, unit, &flushed, &ipacct[0][set]);
}
RWLOCK_EXIT(&ipf_mutex);
return flushed;
}
char *memstr(src, dst, slen, dlen)
char *src, *dst;
int slen, dlen;
{
char *s = NULL;
while (dlen >= slen) {
if (bcmp(src, dst, slen) == 0) {
s = dst;
break;
}
dst++;
dlen--;
}
return s;
}
void fixskip(listp, rp, addremove)
frentry_t **listp, *rp;
int addremove;
{
frentry_t *fp;
int rules = 0, rn = 0;
for (fp = *listp; fp && (fp != rp); fp = fp->fr_next, rules++)
;
if (!fp)
return;
for (fp = *listp; fp && (fp != rp); fp = fp->fr_next, rn++)
if (fp->fr_skip && (rn + fp->fr_skip >= rules))
fp->fr_skip += addremove;
}
#ifdef _KERNEL
/*
* count consecutive 1's in bit mask. If the mask generated by counting
* consecutive 1's is different to that passed, return -1, else return #
* of bits.
*/
int countbits(ip)
u_32_t ip;
{
u_32_t ipn;
int cnt = 0, i, j;
ip = ipn = ntohl(ip);
for (i = 32; i; i--, ipn *= 2)
if (ipn & 0x80000000)
cnt++;
else
break;
ipn = 0;
for (i = 32, j = cnt; i; i--, j--) {
ipn *= 2;
if (j > 0)
ipn++;
}
if (ipn == ip)
return cnt;
return -1;
}
/*
* return the first IP Address associated with an interface
*/
int fr_ifpaddr(v, ifptr, inp)
int v;
void *ifptr;
struct in_addr *inp;
{
# ifdef USE_INET6
struct in6_addr *inp6 = NULL;
# endif
# if SOLARIS
ill_t *ill = ifptr;
# else
struct ifnet *ifp = ifptr;
# endif
struct in_addr in;
# if SOLARIS
# ifdef USE_INET6
if (v == 6) {
struct in6_addr in6;
/*
* First is always link local.
*/
if (ill->ill_ipif->ipif_next)
in6 = ill->ill_ipif->ipif_next->ipif_v6lcl_addr;
else
bzero((char *)&in6, sizeof(in6));
bcopy((char *)&in6, (char *)inp, sizeof(in6));
} else
# endif
{
in.s_addr = ill->ill_ipif->ipif_local_addr;
*inp = in;
}
# else /* SOLARIS */
# if linux
;
# else /* linux */
struct sockaddr_in *sin;
struct ifaddr *ifa;
# if (__FreeBSD_version >= 300000)
ifa = TAILQ_FIRST(&ifp->if_addrhead);
# else
# if defined(__NetBSD__) || defined(__OpenBSD__)
ifa = ifp->if_addrlist.tqh_first;
# else
# if defined(__sgi) && defined(IFF_DRVRLOCK) /* IRIX 6 */
ifa = &((struct in_ifaddr *)ifp->in_ifaddr)->ia_ifa;
# else
ifa = ifp->if_addrlist;
# endif
# endif /* __NetBSD__ || __OpenBSD__ */
# endif /* __FreeBSD_version >= 300000 */
# if (BSD < 199306) && !(/*IRIX6*/defined(__sgi) && defined(IFF_DRVRLOCK))
sin = (struct sockaddr_in *)&ifa->ifa_addr;
# else
sin = (struct sockaddr_in *)ifa->ifa_addr;
while (sin && ifa) {
if ((v == 4) && (sin->sin_family == AF_INET))
break;
# ifdef USE_INET6
if ((v == 6) && (sin->sin_family == AF_INET6)) {
inp6 = &((struct sockaddr_in6 *)sin)->sin6_addr;
if (!IN6_IS_ADDR_LINKLOCAL(inp6) &&
!IN6_IS_ADDR_LOOPBACK(inp6))
break;
}
# endif
# if (__FreeBSD_version >= 300000)
ifa = TAILQ_NEXT(ifa, ifa_link);
# else
# if defined(__NetBSD__) || defined(__OpenBSD__)
ifa = ifa->ifa_list.tqe_next;
# else
ifa = ifa->ifa_next;
# endif
# endif /* __FreeBSD_version >= 300000 */
if (ifa)
sin = (struct sockaddr_in *)ifa->ifa_addr;
}
if (ifa == NULL)
sin = NULL;
if (sin == NULL)
return -1;
# endif /* (BSD < 199306) && (!__sgi && IFF_DRVLOCK) */
# ifdef USE_INET6
if (v == 6)
bcopy((char *)inp6, (char *)inp, sizeof(*inp6));
else
# endif
{
in = sin->sin_addr;
*inp = in;
}
# endif /* linux */
# endif /* SOLARIS */
return 0;
}
static void frsynclist(fr)
frentry_t *fr;
{
for (; fr; fr = fr->fr_next) {
if (fr->fr_ifa != NULL) {
fr->fr_ifa = GETUNIT(fr->fr_ifname, fr->fr_ip.fi_v);
if (fr->fr_ifa == NULL)
fr->fr_ifa = (void *)-1;
}
if (fr->fr_grp)
frsynclist(fr->fr_grp);
}
}
void frsync()
{
# if !SOLARIS
register struct ifnet *ifp;
# if defined(__OpenBSD__) || ((NetBSD >= 199511) && (NetBSD < 1991011)) || \
(defined(__FreeBSD_version) && (__FreeBSD_version >= 300000))
# if (NetBSD >= 199905) || defined(__OpenBSD__)
for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_list.tqe_next)
# else
for (ifp = ifnet.tqh_first; ifp; ifp = ifp->if_link.tqe_next)
# endif
# else
for (ifp = ifnet; ifp; ifp = ifp->if_next)
# endif
{
ip_natsync(ifp);
ip_statesync(ifp);
}
ip_natsync((struct ifnet *)-1);
# endif
WRITE_ENTER(&ipf_mutex);
frsynclist(ipacct[0][fr_active]);
frsynclist(ipacct[1][fr_active]);
frsynclist(ipfilter[0][fr_active]);
frsynclist(ipfilter[1][fr_active]);
#ifdef USE_INET6
frsynclist(ipacct6[0][fr_active]);
frsynclist(ipacct6[1][fr_active]);
frsynclist(ipfilter6[0][fr_active]);
frsynclist(ipfilter6[1][fr_active]);
#endif
RWLOCK_EXIT(&ipf_mutex);
}
/*
* In the functions below, bcopy() is called because the pointer being
* copied _from_ in this instance is a pointer to a char buf (which could
* end up being unaligned) and on the kernel's local stack.
*/
int ircopyptr(a, b, c)
void *a, *b;
size_t c;
{
caddr_t ca;
int err;
#if SOLARIS
if (copyin(a, &ca, sizeof(ca)))
return EFAULT;
#else
bcopy(a, &ca, sizeof(ca));
#endif
err = copyin(ca, b, c);
if (err)
err = EFAULT;
return err;
}
int iwcopyptr(a, b, c)
void *a, *b;
size_t c;
{
caddr_t ca;
int err;
#if SOLARIS
if (copyin(b, &ca, sizeof(ca)))
return EFAULT;
#else
bcopy(b, &ca, sizeof(ca));
#endif
err = copyout(a, ca, c);
if (err)
err = EFAULT;
return err;
}
#else /* _KERNEL */
/*
* return the first IP Address associated with an interface
*/
int fr_ifpaddr(v, ifptr, inp)
int v;
void *ifptr;
struct in_addr *inp;
{
return 0;
}
int ircopyptr(a, b, c)
void *a, *b;
size_t c;
{
caddr_t ca;
bcopy(a, &ca, sizeof(ca));
bcopy(ca, b, c);
return 0;
}
int iwcopyptr(a, b, c)
void *a, *b;
size_t c;
{
caddr_t ca;
bcopy(b, &ca, sizeof(ca));
bcopy(a, ca, c);
return 0;
}
#endif
int fr_lock(data, lockp)
caddr_t data;
int *lockp;
{
int arg, error;
error = IRCOPY(data, (caddr_t)&arg, sizeof(arg));
if (!error) {
error = IWCOPY((caddr_t)lockp, data, sizeof(*lockp));
if (!error)
*lockp = arg;
}
return error;
}
void fr_getstat(fiop)
friostat_t *fiop;
{
bcopy((char *)frstats, (char *)fiop->f_st, sizeof(filterstats_t) * 2);
fiop->f_locks[0] = fr_state_lock;
fiop->f_locks[1] = fr_nat_lock;
fiop->f_locks[2] = fr_frag_lock;
fiop->f_locks[3] = fr_auth_lock;
fiop->f_fin[0] = ipfilter[0][0];
fiop->f_fin[1] = ipfilter[0][1];
fiop->f_fout[0] = ipfilter[1][0];
fiop->f_fout[1] = ipfilter[1][1];
fiop->f_acctin[0] = ipacct[0][0];
fiop->f_acctin[1] = ipacct[0][1];
fiop->f_acctout[0] = ipacct[1][0];
fiop->f_acctout[1] = ipacct[1][1];
#ifdef USE_INET6
fiop->f_fin6[0] = ipfilter6[0][0];
fiop->f_fin6[1] = ipfilter6[0][1];
fiop->f_fout6[0] = ipfilter6[1][0];
fiop->f_fout6[1] = ipfilter6[1][1];
fiop->f_acctin6[0] = ipacct6[0][0];
fiop->f_acctin6[1] = ipacct6[0][1];
fiop->f_acctout6[0] = ipacct6[1][0];
fiop->f_acctout6[1] = ipacct6[1][1];
#endif
fiop->f_active = fr_active;
fiop->f_froute[0] = ipl_frouteok[0];
fiop->f_froute[1] = ipl_frouteok[1];
fiop->f_running = fr_running;
fiop->f_groups[0][0] = ipfgroups[0][0];
fiop->f_groups[0][1] = ipfgroups[0][1];
fiop->f_groups[1][0] = ipfgroups[1][0];
fiop->f_groups[1][1] = ipfgroups[1][1];
fiop->f_groups[2][0] = ipfgroups[2][0];
fiop->f_groups[2][1] = ipfgroups[2][1];
#ifdef IPFILTER_LOG
fiop->f_logging = 1;
#else
fiop->f_logging = 0;
#endif
fiop->f_defpass = fr_pass;
strncpy(fiop->f_version, ipfilter_version, sizeof(fiop->f_version));
}
#ifdef USE_INET6
int icmptoicmp6types[ICMP_MAXTYPE+1] = {
ICMP6_ECHO_REPLY, /* 0: ICMP_ECHOREPLY */
-1, /* 1: UNUSED */
-1, /* 2: UNUSED */
ICMP6_DST_UNREACH, /* 3: ICMP_UNREACH */
-1, /* 4: ICMP_SOURCEQUENCH */
ND_REDIRECT, /* 5: ICMP_REDIRECT */
-1, /* 6: UNUSED */
-1, /* 7: UNUSED */
ICMP6_ECHO_REQUEST, /* 8: ICMP_ECHO */
-1, /* 9: UNUSED */
-1, /* 10: UNUSED */
ICMP6_TIME_EXCEEDED, /* 11: ICMP_TIMXCEED */
ICMP6_PARAM_PROB, /* 12: ICMP_PARAMPROB */
-1, /* 13: ICMP_TSTAMP */
-1, /* 14: ICMP_TSTAMPREPLY */
-1, /* 15: ICMP_IREQ */
-1, /* 16: ICMP_IREQREPLY */
-1, /* 17: ICMP_MASKREQ */
-1, /* 18: ICMP_MASKREPLY */
};
int icmptoicmp6unreach[ICMP_MAX_UNREACH] = {
ICMP6_DST_UNREACH_ADDR, /* 0: ICMP_UNREACH_NET */
ICMP6_DST_UNREACH_ADDR, /* 1: ICMP_UNREACH_HOST */
-1, /* 2: ICMP_UNREACH_PROTOCOL */
ICMP6_DST_UNREACH_NOPORT, /* 3: ICMP_UNREACH_PORT */
-1, /* 4: ICMP_UNREACH_NEEDFRAG */
ICMP6_DST_UNREACH_NOTNEIGHBOR, /* 5: ICMP_UNREACH_SRCFAIL */
ICMP6_DST_UNREACH_ADDR, /* 6: ICMP_UNREACH_NET_UNKNOWN */
ICMP6_DST_UNREACH_ADDR, /* 7: ICMP_UNREACH_HOST_UNKNOWN */
-1, /* 8: ICMP_UNREACH_ISOLATED */
ICMP6_DST_UNREACH_ADMIN, /* 9: ICMP_UNREACH_NET_PROHIB */
ICMP6_DST_UNREACH_ADMIN, /* 10: ICMP_UNREACH_HOST_PROHIB */
-1, /* 11: ICMP_UNREACH_TOSNET */
-1, /* 12: ICMP_UNREACH_TOSHOST */
ICMP6_DST_UNREACH_ADMIN, /* 13: ICMP_UNREACH_ADMIN_PROHIBIT */
};
#endif