version 1.64, 1998/05/01 03:23:24 |
version 1.147, 2002/04/18 22:33:21 |
|
|
/* $NetBSD$ */ |
/* $NetBSD$ */ |
|
|
|
/* |
|
* 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) 1998 The NetBSD Foundation, Inc. |
* Copyright (c) 1998 The NetBSD Foundation, Inc. |
* All rights reserved. |
* All rights reserved. |
|
|
* @(#)ip_input.c 8.2 (Berkeley) 1/4/94 |
* @(#)ip_input.c 8.2 (Berkeley) 1/4/94 |
*/ |
*/ |
|
|
|
#include <sys/cdefs.h> |
|
__KERNEL_RCSID(0, "$NetBSD$"); |
|
|
#include "opt_gateway.h" |
#include "opt_gateway.h" |
|
#include "opt_pfil_hooks.h" |
|
#include "opt_ipsec.h" |
#include "opt_mrouting.h" |
#include "opt_mrouting.h" |
|
#include "opt_inet_csum.h" |
|
|
#include <sys/param.h> |
#include <sys/param.h> |
#include <sys/systm.h> |
#include <sys/systm.h> |
|
|
#include <sys/errno.h> |
#include <sys/errno.h> |
#include <sys/time.h> |
#include <sys/time.h> |
#include <sys/kernel.h> |
#include <sys/kernel.h> |
#include <sys/proc.h> |
#include <sys/pool.h> |
|
|
#include <vm/vm.h> |
|
#include <sys/sysctl.h> |
#include <sys/sysctl.h> |
|
|
#include <net/if.h> |
#include <net/if.h> |
|
|
#include <netinet/in_var.h> |
#include <netinet/in_var.h> |
#include <netinet/ip_var.h> |
#include <netinet/ip_var.h> |
#include <netinet/ip_icmp.h> |
#include <netinet/ip_icmp.h> |
|
/* just for gif_ttl */ |
|
#include <netinet/in_gif.h> |
|
#include "gif.h" |
|
#include <net/if_gre.h> |
|
#include "gre.h" |
|
|
|
#ifdef MROUTING |
|
#include <netinet/ip_mroute.h> |
|
#endif |
|
|
|
#ifdef IPSEC |
|
#include <netinet6/ipsec.h> |
|
#include <netkey/key.h> |
|
#endif |
|
|
#ifndef IPFORWARDING |
#ifndef IPFORWARDING |
#ifdef GATEWAY |
#ifdef GATEWAY |
Line 153 int ipprintfs = 0; |
|
Line 200 int ipprintfs = 0; |
|
struct rttimer_queue *ip_mtudisc_timeout_q = NULL; |
struct rttimer_queue *ip_mtudisc_timeout_q = NULL; |
|
|
extern struct domain inetdomain; |
extern struct domain inetdomain; |
extern struct protosw inetsw[]; |
|
u_char ip_protox[IPPROTO_MAX]; |
|
int ipqmaxlen = IFQ_MAXLEN; |
int ipqmaxlen = IFQ_MAXLEN; |
struct in_ifaddrhead in_ifaddr; |
struct in_ifaddrhead in_ifaddr; |
struct in_ifaddrhashhead *in_ifaddrhashtbl; |
struct in_ifaddrhashhead *in_ifaddrhashtbl; |
struct ifqueue ipintrq; |
struct ifqueue ipintrq; |
struct ipstat ipstat; |
struct ipstat ipstat; |
u_int16_t ip_id; |
u_int16_t ip_id; |
int ip_defttl; |
|
|
#ifdef PFIL_HOOKS |
|
struct pfil_head inet_pfil_hook; |
|
#endif |
|
|
struct ipqhead ipq; |
struct ipqhead ipq; |
|
int ipq_locked; |
|
int ip_nfragpackets = 0; |
|
int ip_maxfragpackets = 200; |
|
|
|
static __inline int ipq_lock_try __P((void)); |
|
static __inline void ipq_unlock __P((void)); |
|
|
|
static __inline int |
|
ipq_lock_try() |
|
{ |
|
int s; |
|
|
|
/* |
|
* Use splvm() -- we're bloking things that would cause |
|
* mbuf allocation. |
|
*/ |
|
s = splvm(); |
|
if (ipq_locked) { |
|
splx(s); |
|
return (0); |
|
} |
|
ipq_locked = 1; |
|
splx(s); |
|
return (1); |
|
} |
|
|
|
static __inline void |
|
ipq_unlock() |
|
{ |
|
int s; |
|
|
|
s = splvm(); |
|
ipq_locked = 0; |
|
splx(s); |
|
} |
|
|
|
#ifdef DIAGNOSTIC |
|
#define IPQ_LOCK() \ |
|
do { \ |
|
if (ipq_lock_try() == 0) { \ |
|
printf("%s:%d: ipq already locked\n", __FILE__, __LINE__); \ |
|
panic("ipq_lock"); \ |
|
} \ |
|
} while (0) |
|
#define IPQ_LOCK_CHECK() \ |
|
do { \ |
|
if (ipq_locked == 0) { \ |
|
printf("%s:%d: ipq lock not held\n", __FILE__, __LINE__); \ |
|
panic("ipq lock check"); \ |
|
} \ |
|
} while (0) |
|
#else |
|
#define IPQ_LOCK() (void) ipq_lock_try() |
|
#define IPQ_LOCK_CHECK() /* nothing */ |
|
#endif |
|
|
|
#define IPQ_UNLOCK() ipq_unlock() |
|
|
|
struct pool ipqent_pool; |
|
|
|
#ifdef INET_CSUM_COUNTERS |
|
#include <sys/device.h> |
|
|
|
struct evcnt ip_hwcsum_bad = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, |
|
NULL, "inet", "hwcsum bad"); |
|
struct evcnt ip_hwcsum_ok = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, |
|
NULL, "inet", "hwcsum ok"); |
|
struct evcnt ip_swcsum = EVCNT_INITIALIZER(EVCNT_TYPE_MISC, |
|
NULL, "inet", "swcsum"); |
|
|
|
#define INET_CSUM_COUNTER_INCR(ev) (ev)->ev_count++ |
|
|
|
#else |
|
|
|
#define INET_CSUM_COUNTER_INCR(ev) /* nothing */ |
|
|
|
#endif /* INET_CSUM_COUNTERS */ |
|
|
/* |
/* |
* We need to save the IP options in case a protocol wants to respond |
* We need to save the IP options in case a protocol wants to respond |
Line 188 static void save_rte __P((u_char *, stru |
|
Line 314 static void save_rte __P((u_char *, stru |
|
void |
void |
ip_init() |
ip_init() |
{ |
{ |
register struct protosw *pr; |
struct protosw *pr; |
register int i; |
int i; |
|
|
|
pool_init(&ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", |
|
NULL); |
|
|
pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); |
pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); |
if (pr == 0) |
if (pr == 0) |
|
|
ip_id = time.tv_sec & 0xffff; |
ip_id = time.tv_sec & 0xffff; |
ipintrq.ifq_maxlen = ipqmaxlen; |
ipintrq.ifq_maxlen = ipqmaxlen; |
TAILQ_INIT(&in_ifaddr); |
TAILQ_INIT(&in_ifaddr); |
in_ifaddrhashtbl = |
in_ifaddrhashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IFADDR, |
hashinit(IN_IFADDR_HASH_SIZE, M_IFADDR, M_WAITOK, &in_ifaddrhash); |
M_WAITOK, &in_ifaddrhash); |
if (ip_mtudisc != 0) |
if (ip_mtudisc != 0) |
ip_mtudisc_timeout_q = |
ip_mtudisc_timeout_q = |
rt_timer_queue_create(ip_mtudisc_timeout); |
rt_timer_queue_create(ip_mtudisc_timeout); |
|
#ifdef GATEWAY |
|
ipflow_init(); |
|
#endif |
|
|
|
#ifdef PFIL_HOOKS |
|
/* Register our Packet Filter hook. */ |
|
inet_pfil_hook.ph_type = PFIL_TYPE_AF; |
|
inet_pfil_hook.ph_af = AF_INET; |
|
i = pfil_head_register(&inet_pfil_hook); |
|
if (i != 0) |
|
printf("ip_init: WARNING: unable to register pfil hook, " |
|
"error %d\n", i); |
|
#endif /* PFIL_HOOKS */ |
|
|
|
#ifdef INET_CSUM_COUNTERS |
|
evcnt_attach_static(&ip_hwcsum_bad); |
|
evcnt_attach_static(&ip_hwcsum_ok); |
|
evcnt_attach_static(&ip_swcsum); |
|
#endif /* INET_CSUM_COUNTERS */ |
} |
} |
|
|
struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; |
struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; |
struct route ipforward_rt; |
struct route ipforward_rt; |
|
|
/* |
/* |
|
* IP software interrupt routine |
|
*/ |
|
void |
|
ipintr() |
|
{ |
|
int s; |
|
struct mbuf *m; |
|
|
|
while (1) { |
|
s = splnet(); |
|
IF_DEQUEUE(&ipintrq, m); |
|
splx(s); |
|
if (m == 0) |
|
return; |
|
ip_input(m); |
|
} |
|
} |
|
|
|
/* |
* Ip input routine. Checksum and byte swap header. If fragmented |
* Ip input routine. Checksum and byte swap header. If fragmented |
* try to reassemble. Process options. Pass to next level. |
* try to reassemble. Process options. Pass to next level. |
*/ |
*/ |
void |
void |
ipintr() |
ip_input(struct mbuf *m) |
{ |
{ |
register struct ip *ip = NULL; |
struct ip *ip = NULL; |
register struct mbuf *m; |
struct ipq *fp; |
register struct ipq *fp; |
struct in_ifaddr *ia; |
register struct in_ifaddr *ia; |
struct ifaddr *ifa; |
register struct ifaddr *ifa; |
|
struct ipqent *ipqe; |
struct ipqent *ipqe; |
int hlen = 0, mff, len, s; |
int hlen = 0, mff, len; |
#ifdef PFIL_HOOKS |
int downmatch; |
struct packet_filter_hook *pfh; |
|
struct mbuf *m0; |
|
int rv; |
|
#endif /* PFIL_HOOKS */ |
|
|
|
next: |
|
/* |
|
* Get next datagram off input queue and get IP header |
|
* in first mbuf. |
|
*/ |
|
s = splimp(); |
|
IF_DEQUEUE(&ipintrq, m); |
|
splx(s); |
|
if (m == 0) |
|
return; |
|
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if ((m->m_flags & M_PKTHDR) == 0) |
if ((m->m_flags & M_PKTHDR) == 0) |
panic("ipintr no HDR"); |
panic("ipintr no HDR"); |
#endif |
#endif |
|
#ifdef IPSEC |
|
/* |
|
* should the inner packet be considered authentic? |
|
* see comment in ah4_input(). |
|
*/ |
|
if (m) { |
|
m->m_flags &= ~M_AUTHIPHDR; |
|
m->m_flags &= ~M_AUTHIPDGM; |
|
} |
|
#endif |
/* |
/* |
* If no IP addresses have been set yet but the interfaces |
* If no IP addresses have been set yet but the interfaces |
* are receiving, can't do anything with incoming packets yet. |
* are receiving, can't do anything with incoming packets yet. |
*/ |
*/ |
if (in_ifaddr.tqh_first == 0) |
if (TAILQ_FIRST(&in_ifaddr) == 0) |
goto bad; |
goto bad; |
ipstat.ips_total++; |
ipstat.ips_total++; |
if (m->m_len < sizeof (struct ip) && |
if (m->m_len < sizeof (struct ip) && |
(m = m_pullup(m, sizeof (struct ip))) == 0) { |
(m = m_pullup(m, sizeof (struct ip))) == 0) { |
ipstat.ips_toosmall++; |
ipstat.ips_toosmall++; |
goto next; |
return; |
} |
} |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
if (ip->ip_v != IPVERSION) { |
if (ip->ip_v != IPVERSION) { |
|
|
if (hlen > m->m_len) { |
if (hlen > m->m_len) { |
if ((m = m_pullup(m, hlen)) == 0) { |
if ((m = m_pullup(m, hlen)) == 0) { |
ipstat.ips_badhlen++; |
ipstat.ips_badhlen++; |
goto next; |
return; |
} |
} |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
} |
} |
if ((ip->ip_sum = in_cksum(m, hlen)) != 0) { |
|
ipstat.ips_badsum++; |
/* |
|
* RFC1122: packets with a multicast source address are |
|
* not allowed. |
|
*/ |
|
if (IN_MULTICAST(ip->ip_src.s_addr)) { |
|
ipstat.ips_badaddr++; |
goto bad; |
goto bad; |
} |
} |
|
|
|
/* 127/8 must not appear on wire - RFC1122 */ |
|
if ((ntohl(ip->ip_dst.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET || |
|
(ntohl(ip->ip_src.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET) { |
|
if ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0) { |
|
ipstat.ips_badaddr++; |
|
goto bad; |
|
} |
|
} |
|
|
|
switch (m->m_pkthdr.csum_flags & |
|
((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_IPv4) | |
|
M_CSUM_IPv4_BAD)) { |
|
case M_CSUM_IPv4|M_CSUM_IPv4_BAD: |
|
INET_CSUM_COUNTER_INCR(&ip_hwcsum_bad); |
|
goto badcsum; |
|
|
|
case M_CSUM_IPv4: |
|
/* Checksum was okay. */ |
|
INET_CSUM_COUNTER_INCR(&ip_hwcsum_ok); |
|
break; |
|
|
|
default: |
|
/* Must compute it ourselves. */ |
|
INET_CSUM_COUNTER_INCR(&ip_swcsum); |
|
if (in_cksum(m, hlen) != 0) |
|
goto bad; |
|
break; |
|
} |
|
|
|
/* Retrieve the packet length. */ |
|
len = ntohs(ip->ip_len); |
|
|
/* |
/* |
* Convert fields to host representation. |
* Check for additional length bogosity |
*/ |
*/ |
NTOHS(ip->ip_len); |
if (len < hlen) { |
NTOHS(ip->ip_id); |
ipstat.ips_badlen++; |
NTOHS(ip->ip_off); |
goto bad; |
len = ip->ip_len; |
} |
|
|
/* |
/* |
* Check that the amount of data in the buffers |
* Check that the amount of data in the buffers |
|
|
m_adj(m, len - m->m_pkthdr.len); |
m_adj(m, len - m->m_pkthdr.len); |
} |
} |
|
|
|
#ifdef IPSEC |
|
/* ipflow (IP fast fowarding) is not compatible with IPsec. */ |
|
m->m_flags &= ~M_CANFASTFWD; |
|
#else |
/* |
/* |
* Assume that we can create a fast-forward IP flow entry |
* Assume that we can create a fast-forward IP flow entry |
* based on this packet. |
* based on this packet. |
*/ |
*/ |
m->m_flags |= M_CANFASTFWD; |
m->m_flags |= M_CANFASTFWD; |
|
#endif |
|
|
#ifdef PFIL_HOOKS |
#ifdef PFIL_HOOKS |
/* |
/* |
|
|
* Note that filters must _never_ set this flag, as another filter |
* Note that filters must _never_ set this flag, as another filter |
* in the list may have previously cleared it. |
* in the list may have previously cleared it. |
*/ |
*/ |
m0 = m; |
/* |
for (pfh = pfil_hook_get(PFIL_IN); pfh; pfh = pfh->pfil_link.tqe_next) |
* let ipfilter look at packet on the wire, |
if (pfh->pfil_func) { |
* not the decapsulated packet. |
rv = pfh->pfil_func(ip, hlen, m->m_pkthdr.rcvif, 0, &m0); |
*/ |
if (rv) |
#ifdef IPSEC |
goto next; |
if (!ipsec_getnhist(m)) |
ip = mtod(m = m0, struct ip *); |
#else |
} |
if (1) |
|
#endif |
|
{ |
|
if (pfil_run_hooks(&inet_pfil_hook, &m, m->m_pkthdr.rcvif, |
|
PFIL_IN) != 0) |
|
return; |
|
if (m == NULL) |
|
return; |
|
ip = mtod(m, struct ip *); |
|
hlen = ip->ip_hl << 2; |
|
} |
#endif /* PFIL_HOOKS */ |
#endif /* PFIL_HOOKS */ |
|
|
|
#ifdef ALTQ |
|
/* XXX Temporary until ALTQ is changed to use a pfil hook */ |
|
if (altq_input != NULL && (*altq_input)(m, AF_INET) == 0) { |
|
/* packet dropped by traffic conditioner */ |
|
return; |
|
} |
|
#endif |
|
|
|
/* |
|
* Convert fields to host representation. |
|
*/ |
|
NTOHS(ip->ip_len); |
|
NTOHS(ip->ip_off); |
|
|
/* |
/* |
* Process options and, if not destined for us, |
* Process options and, if not destined for us, |
* ship it on. ip_dooptions returns 1 when an |
* ship it on. ip_dooptions returns 1 when an |
|
|
*/ |
*/ |
ip_nhops = 0; /* for source routed packets */ |
ip_nhops = 0; /* for source routed packets */ |
if (hlen > sizeof (struct ip) && ip_dooptions(m)) |
if (hlen > sizeof (struct ip) && ip_dooptions(m)) |
goto next; |
return; |
|
|
/* |
/* |
* Check our list of addresses, to see if the packet is for us. |
* Check our list of addresses, to see if the packet is for us. |
*/ |
* |
INADDR_TO_IA(ip->ip_dst, ia); |
* Traditional 4.4BSD did not consult IFF_UP at all. |
if (ia != NULL) goto ours; |
* The behavior here is to treat addresses on !IFF_UP interface |
|
* as not mine. |
|
*/ |
|
downmatch = 0; |
|
LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { |
|
if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { |
|
if ((ia->ia_ifp->if_flags & IFF_UP) != 0) |
|
break; |
|
else |
|
downmatch++; |
|
} |
|
} |
|
if (ia != NULL) |
|
goto ours; |
if (m->m_pkthdr.rcvif->if_flags & IFF_BROADCAST) { |
if (m->m_pkthdr.rcvif->if_flags & IFF_BROADCAST) { |
for (ifa = m->m_pkthdr.rcvif->if_addrlist.tqh_first; |
TAILQ_FOREACH(ifa, &m->m_pkthdr.rcvif->if_addrlist, ifa_list) { |
ifa != NULL; ifa = ifa->ifa_list.tqe_next) { |
if (ifa->ifa_addr->sa_family != AF_INET) |
if (ifa->ifa_addr->sa_family != AF_INET) continue; |
continue; |
ia = ifatoia(ifa); |
ia = ifatoia(ifa); |
if (in_hosteq(ip->ip_dst, ia->ia_broadaddr.sin_addr) || |
if (in_hosteq(ip->ip_dst, ia->ia_broadaddr.sin_addr) || |
in_hosteq(ip->ip_dst, ia->ia_netbroadcast) || |
in_hosteq(ip->ip_dst, ia->ia_netbroadcast) || |
|
|
#ifdef MROUTING |
#ifdef MROUTING |
extern struct socket *ip_mrouter; |
extern struct socket *ip_mrouter; |
|
|
if (m->m_flags & M_EXT) { |
if (M_READONLY(m)) { |
if ((m = m_pullup(m, hlen)) == 0) { |
if ((m = m_pullup(m, hlen)) == 0) { |
ipstat.ips_toosmall++; |
ipstat.ips_toosmall++; |
goto next; |
return; |
} |
} |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
} |
} |
|
|
* as expected when ip_mforward() is called from |
* as expected when ip_mforward() is called from |
* ip_output().) |
* ip_output().) |
*/ |
*/ |
ip->ip_id = htons(ip->ip_id); |
|
if (ip_mforward(m, m->m_pkthdr.rcvif) != 0) { |
if (ip_mforward(m, m->m_pkthdr.rcvif) != 0) { |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
m_freem(m); |
m_freem(m); |
goto next; |
return; |
} |
} |
ip->ip_id = ntohs(ip->ip_id); |
|
|
|
/* |
/* |
* The process-level routing demon needs to receive |
* The process-level routing demon needs to receive |
|
|
if (inm == NULL) { |
if (inm == NULL) { |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
m_freem(m); |
m_freem(m); |
goto next; |
return; |
} |
} |
goto ours; |
goto ours; |
} |
} |
|
|
if (ipforwarding == 0) { |
if (ipforwarding == 0) { |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
m_freem(m); |
m_freem(m); |
} else |
} else { |
|
/* |
|
* If ip_dst matched any of my address on !IFF_UP interface, |
|
* and there's no IFF_UP interface that matches ip_dst, |
|
* send icmp unreach. Forwarding it will result in in-kernel |
|
* forwarding loop till TTL goes to 0. |
|
*/ |
|
if (downmatch) { |
|
icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, 0, 0); |
|
ipstat.ips_cantforward++; |
|
return; |
|
} |
|
#ifdef IPSEC |
|
if (ipsec4_in_reject(m, NULL)) { |
|
ipsecstat.in_polvio++; |
|
goto bad; |
|
} |
|
#endif |
|
|
ip_forward(m, 0); |
ip_forward(m, 0); |
goto next; |
} |
|
return; |
|
|
ours: |
ours: |
/* |
/* |
|
|
* Look for queue of fragments |
* Look for queue of fragments |
* of this datagram. |
* of this datagram. |
*/ |
*/ |
for (fp = ipq.lh_first; fp != NULL; fp = fp->ipq_q.le_next) |
IPQ_LOCK(); |
|
LIST_FOREACH(fp, &ipq, ipq_q) |
if (ip->ip_id == fp->ipq_id && |
if (ip->ip_id == fp->ipq_id && |
in_hosteq(ip->ip_src, fp->ipq_src) && |
in_hosteq(ip->ip_src, fp->ipq_src) && |
in_hosteq(ip->ip_dst, fp->ipq_dst) && |
in_hosteq(ip->ip_dst, fp->ipq_dst) && |
|
|
*/ |
*/ |
if (ip->ip_len == 0 || (ip->ip_len & 0x7) != 0) { |
if (ip->ip_len == 0 || (ip->ip_len & 0x7) != 0) { |
ipstat.ips_badfrags++; |
ipstat.ips_badfrags++; |
|
IPQ_UNLOCK(); |
goto bad; |
goto bad; |
} |
} |
} |
} |
|
|
*/ |
*/ |
if (mff || ip->ip_off) { |
if (mff || ip->ip_off) { |
ipstat.ips_fragments++; |
ipstat.ips_fragments++; |
MALLOC(ipqe, struct ipqent *, sizeof (struct ipqent), |
ipqe = pool_get(&ipqent_pool, PR_NOWAIT); |
M_IPQ, M_NOWAIT); |
|
if (ipqe == NULL) { |
if (ipqe == NULL) { |
ipstat.ips_rcvmemdrop++; |
ipstat.ips_rcvmemdrop++; |
|
IPQ_UNLOCK(); |
goto bad; |
goto bad; |
} |
} |
ipqe->ipqe_mff = mff; |
ipqe->ipqe_mff = mff; |
ipqe->ipqe_m = m; |
ipqe->ipqe_m = m; |
ipqe->ipqe_ip = ip; |
ipqe->ipqe_ip = ip; |
m = ip_reass(ipqe, fp); |
m = ip_reass(ipqe, fp); |
if (m == 0) |
if (m == 0) { |
goto next; |
IPQ_UNLOCK(); |
|
return; |
|
} |
ipstat.ips_reassembled++; |
ipstat.ips_reassembled++; |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
|
hlen = ip->ip_hl << 2; |
|
ip->ip_len += hlen; |
} else |
} else |
if (fp) |
if (fp) |
ip_freef(fp); |
ip_freef(fp); |
} else |
IPQ_UNLOCK(); |
ip->ip_len -= hlen; |
} |
|
|
|
#ifdef IPSEC |
|
/* |
|
* enforce IPsec policy checking if we are seeing last header. |
|
* note that we do not visit this with protocols with pcb layer |
|
* code - like udp/tcp/raw ip. |
|
*/ |
|
if ((inetsw[ip_protox[ip->ip_p]].pr_flags & PR_LASTHDR) != 0 && |
|
ipsec4_in_reject(m, NULL)) { |
|
ipsecstat.in_polvio++; |
|
goto bad; |
|
} |
|
#endif |
|
|
/* |
/* |
* Switch out to protocol's input routine. |
* Switch out to protocol's input routine. |
*/ |
*/ |
|
#if IFA_STATS |
|
if (ia && ip) |
|
ia->ia_ifa.ifa_data.ifad_inbytes += ip->ip_len; |
|
#endif |
ipstat.ips_delivered++; |
ipstat.ips_delivered++; |
(*inetsw[ip_protox[ip->ip_p]].pr_input)(m, hlen); |
{ |
goto next; |
int off = hlen, nh = ip->ip_p; |
|
|
|
(*inetsw[ip_protox[nh]].pr_input)(m, off, nh); |
|
return; |
|
} |
bad: |
bad: |
m_freem(m); |
m_freem(m); |
goto next; |
return; |
|
|
|
badcsum: |
|
ipstat.ips_badsum++; |
|
m_freem(m); |
} |
} |
|
|
/* |
/* |
|
|
*/ |
*/ |
struct mbuf * |
struct mbuf * |
ip_reass(ipqe, fp) |
ip_reass(ipqe, fp) |
register struct ipqent *ipqe; |
struct ipqent *ipqe; |
register struct ipq *fp; |
struct ipq *fp; |
{ |
{ |
register struct mbuf *m = ipqe->ipqe_m; |
struct mbuf *m = ipqe->ipqe_m; |
register struct ipqent *nq, *p, *q; |
struct ipqent *nq, *p, *q; |
struct ip *ip; |
struct ip *ip; |
struct mbuf *t; |
struct mbuf *t; |
int hlen = ipqe->ipqe_ip->ip_hl << 2; |
int hlen = ipqe->ipqe_ip->ip_hl << 2; |
int i, next; |
int i, next; |
|
|
|
IPQ_LOCK_CHECK(); |
|
|
/* |
/* |
* Presence of header sizes in mbufs |
* Presence of header sizes in mbufs |
* would confuse code below. |
* would confuse code below. |
Line 548 ip_reass(ipqe, fp) |
|
Line 839 ip_reass(ipqe, fp) |
|
* If first fragment to arrive, create a reassembly queue. |
* If first fragment to arrive, create a reassembly queue. |
*/ |
*/ |
if (fp == 0) { |
if (fp == 0) { |
|
/* |
|
* Enforce upper bound on number of fragmented packets |
|
* for which we attempt reassembly; |
|
* If maxfrag is 0, never accept fragments. |
|
* If maxfrag is -1, accept all fragments without limitation. |
|
*/ |
|
if (ip_maxfragpackets < 0) |
|
; |
|
else if (ip_nfragpackets >= ip_maxfragpackets) |
|
goto dropfrag; |
|
ip_nfragpackets++; |
MALLOC(fp, struct ipq *, sizeof (struct ipq), |
MALLOC(fp, struct ipq *, sizeof (struct ipq), |
M_FTABLE, M_NOWAIT); |
M_FTABLE, M_NOWAIT); |
if (fp == NULL) |
if (fp == NULL) |
Line 566 ip_reass(ipqe, fp) |
|
Line 868 ip_reass(ipqe, fp) |
|
/* |
/* |
* Find a segment which begins after this one does. |
* Find a segment which begins after this one does. |
*/ |
*/ |
for (p = NULL, q = fp->ipq_fragq.lh_first; q != NULL; |
for (p = NULL, q = LIST_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = q->ipqe_q.le_next) |
p = q, q = LIST_NEXT(q, ipqe_q)) |
if (q->ipqe_ip->ip_off > ipqe->ipqe_ip->ip_off) |
if (q->ipqe_ip->ip_off > ipqe->ipqe_ip->ip_off) |
break; |
break; |
|
|
Line 602 ip_reass(ipqe, fp) |
|
Line 904 ip_reass(ipqe, fp) |
|
m_adj(q->ipqe_m, i); |
m_adj(q->ipqe_m, i); |
break; |
break; |
} |
} |
nq = q->ipqe_q.le_next; |
nq = LIST_NEXT(q, ipqe_q); |
m_freem(q->ipqe_m); |
m_freem(q->ipqe_m); |
LIST_REMOVE(q, ipqe_q); |
LIST_REMOVE(q, ipqe_q); |
FREE(q, M_IPQ); |
pool_put(&ipqent_pool, q); |
} |
} |
|
|
insert: |
insert: |
|
|
LIST_INSERT_AFTER(p, ipqe, ipqe_q); |
LIST_INSERT_AFTER(p, ipqe, ipqe_q); |
} |
} |
next = 0; |
next = 0; |
for (p = NULL, q = fp->ipq_fragq.lh_first; q != NULL; |
for (p = NULL, q = LIST_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = q->ipqe_q.le_next) { |
p = q, q = LIST_NEXT(q, ipqe_q)) { |
if (q->ipqe_ip->ip_off != next) |
if (q->ipqe_ip->ip_off != next) |
return (0); |
return (0); |
next += q->ipqe_ip->ip_len; |
next += q->ipqe_ip->ip_len; |
|
|
* Reassembly is complete. Check for a bogus message size and |
* Reassembly is complete. Check for a bogus message size and |
* concatenate fragments. |
* concatenate fragments. |
*/ |
*/ |
q = fp->ipq_fragq.lh_first; |
q = LIST_FIRST(&fp->ipq_fragq); |
ip = q->ipqe_ip; |
ip = q->ipqe_ip; |
if ((next + (ip->ip_hl << 2)) > IP_MAXPACKET) { |
if ((next + (ip->ip_hl << 2)) > IP_MAXPACKET) { |
ipstat.ips_toolong++; |
ipstat.ips_toolong++; |
|
|
t = m->m_next; |
t = m->m_next; |
m->m_next = 0; |
m->m_next = 0; |
m_cat(m, t); |
m_cat(m, t); |
nq = q->ipqe_q.le_next; |
nq = LIST_NEXT(q, ipqe_q); |
FREE(q, M_IPQ); |
pool_put(&ipqent_pool, q); |
for (q = nq; q != NULL; q = nq) { |
for (q = nq; q != NULL; q = nq) { |
t = q->ipqe_m; |
t = q->ipqe_m; |
nq = q->ipqe_q.le_next; |
nq = LIST_NEXT(q, ipqe_q); |
FREE(q, M_IPQ); |
pool_put(&ipqent_pool, q); |
m_cat(m, t); |
m_cat(m, t); |
} |
} |
|
|
|
|
ip->ip_dst = fp->ipq_dst; |
ip->ip_dst = fp->ipq_dst; |
LIST_REMOVE(fp, ipq_q); |
LIST_REMOVE(fp, ipq_q); |
FREE(fp, M_FTABLE); |
FREE(fp, M_FTABLE); |
|
ip_nfragpackets--; |
m->m_len += (ip->ip_hl << 2); |
m->m_len += (ip->ip_hl << 2); |
m->m_data -= (ip->ip_hl << 2); |
m->m_data -= (ip->ip_hl << 2); |
/* some debugging cruft by sklower, below, will go away soon */ |
/* some debugging cruft by sklower, below, will go away soon */ |
if (m->m_flags & M_PKTHDR) { /* XXX this should be done elsewhere */ |
if (m->m_flags & M_PKTHDR) { /* XXX this should be done elsewhere */ |
register int plen = 0; |
int plen = 0; |
for (t = m; t; t = t->m_next) |
for (t = m; t; t = t->m_next) |
plen += t->m_len; |
plen += t->m_len; |
m->m_pkthdr.len = plen; |
m->m_pkthdr.len = plen; |
|
|
dropfrag: |
dropfrag: |
ipstat.ips_fragdropped++; |
ipstat.ips_fragdropped++; |
m_freem(m); |
m_freem(m); |
FREE(ipqe, M_IPQ); |
pool_put(&ipqent_pool, ipqe); |
return (0); |
return (0); |
} |
} |
|
|
|
|
ip_freef(fp) |
ip_freef(fp) |
struct ipq *fp; |
struct ipq *fp; |
{ |
{ |
register struct ipqent *q, *p; |
struct ipqent *q, *p; |
|
|
|
IPQ_LOCK_CHECK(); |
|
|
for (q = fp->ipq_fragq.lh_first; q != NULL; q = p) { |
for (q = LIST_FIRST(&fp->ipq_fragq); q != NULL; q = p) { |
p = q->ipqe_q.le_next; |
p = LIST_NEXT(q, ipqe_q); |
m_freem(q->ipqe_m); |
m_freem(q->ipqe_m); |
LIST_REMOVE(q, ipqe_q); |
LIST_REMOVE(q, ipqe_q); |
FREE(q, M_IPQ); |
pool_put(&ipqent_pool, q); |
} |
} |
LIST_REMOVE(fp, ipq_q); |
LIST_REMOVE(fp, ipq_q); |
FREE(fp, M_FTABLE); |
FREE(fp, M_FTABLE); |
|
ip_nfragpackets--; |
} |
} |
|
|
/* |
/* |
|
|
void |
void |
ip_slowtimo() |
ip_slowtimo() |
{ |
{ |
register struct ipq *fp, *nfp; |
struct ipq *fp, *nfp; |
int s = splsoftnet(); |
int s = splsoftnet(); |
|
|
for (fp = ipq.lh_first; fp != NULL; fp = nfp) { |
IPQ_LOCK(); |
nfp = fp->ipq_q.le_next; |
for (fp = LIST_FIRST(&ipq); fp != NULL; fp = nfp) { |
|
nfp = LIST_NEXT(fp, ipq_q); |
if (--fp->ipq_ttl == 0) { |
if (--fp->ipq_ttl == 0) { |
ipstat.ips_fragtimeout++; |
ipstat.ips_fragtimeout++; |
ip_freef(fp); |
ip_freef(fp); |
} |
} |
} |
} |
|
/* |
|
* If we are over the maximum number of fragments |
|
* (due to the limit being lowered), drain off |
|
* enough to get down to the new limit. |
|
*/ |
|
if (ip_maxfragpackets < 0) |
|
; |
|
else { |
|
while (ip_nfragpackets > ip_maxfragpackets && LIST_FIRST(&ipq)) |
|
ip_freef(LIST_FIRST(&ipq)); |
|
} |
|
IPQ_UNLOCK(); |
#ifdef GATEWAY |
#ifdef GATEWAY |
ipflow_slowtimo(); |
ipflow_slowtimo(); |
#endif |
#endif |
|
|
ip_drain() |
ip_drain() |
{ |
{ |
|
|
while (ipq.lh_first != NULL) { |
/* |
|
* We may be called from a device's interrupt context. If |
|
* the ipq is already busy, just bail out now. |
|
*/ |
|
if (ipq_lock_try() == 0) |
|
return; |
|
|
|
while (LIST_FIRST(&ipq) != NULL) { |
ipstat.ips_fragdropped++; |
ipstat.ips_fragdropped++; |
ip_freef(ipq.lh_first); |
ip_freef(LIST_FIRST(&ipq)); |
} |
} |
|
|
|
IPQ_UNLOCK(); |
} |
} |
|
|
/* |
/* |
|
|
ip_dooptions(m) |
ip_dooptions(m) |
struct mbuf *m; |
struct mbuf *m; |
{ |
{ |
register struct ip *ip = mtod(m, struct ip *); |
struct ip *ip = mtod(m, struct ip *); |
register u_char *cp; |
u_char *cp, *cp0; |
register struct ip_timestamp *ipt; |
struct ip_timestamp *ipt; |
register struct in_ifaddr *ia; |
struct in_ifaddr *ia; |
int opt, optlen, cnt, off, code, type = ICMP_PARAMPROB, forward = 0; |
int opt, optlen, cnt, off, code, type = ICMP_PARAMPROB, forward = 0; |
struct in_addr *sin, dst; |
struct in_addr dst; |
n_time ntime; |
n_time ntime; |
|
|
dst = ip->ip_dst; |
dst = ip->ip_dst; |
|
Line 1095 ip_dooptions(m) |
|
if (opt == IPOPT_NOP) |
if (opt == IPOPT_NOP) |
optlen = 1; |
optlen = 1; |
else { |
else { |
|
if (cnt < IPOPT_OLEN + sizeof(*cp)) { |
|
code = &cp[IPOPT_OLEN] - (u_char *)ip; |
|
goto bad; |
|
} |
optlen = cp[IPOPT_OLEN]; |
optlen = cp[IPOPT_OLEN]; |
if (optlen <= 0 || optlen > cnt) { |
if (optlen < IPOPT_OLEN + sizeof(*cp) || optlen > cnt) { |
code = &cp[IPOPT_OLEN] - (u_char *)ip; |
code = &cp[IPOPT_OLEN] - (u_char *)ip; |
goto bad; |
goto bad; |
} |
} |
|
Line 1126 ip_dooptions(m) |
|
code = ICMP_UNREACH_NET_PROHIB; |
code = ICMP_UNREACH_NET_PROHIB; |
goto bad; |
goto bad; |
} |
} |
|
if (optlen < IPOPT_OFFSET + sizeof(*cp)) { |
|
code = &cp[IPOPT_OLEN] - (u_char *)ip; |
|
goto bad; |
|
} |
if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { |
if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { |
code = &cp[IPOPT_OFFSET] - (u_char *)ip; |
code = &cp[IPOPT_OFFSET] - (u_char *)ip; |
goto bad; |
goto bad; |
|
Line 1149 ip_dooptions(m) |
|
break; |
break; |
} |
} |
off--; /* 0 origin */ |
off--; /* 0 origin */ |
if (off > optlen - sizeof(struct in_addr)) { |
if ((off + sizeof(struct in_addr)) > optlen) { |
/* |
/* |
* End of source route. Should be for us. |
* End of source route. Should be for us. |
*/ |
*/ |
|
Line 1161 ip_dooptions(m) |
|
*/ |
*/ |
bcopy((caddr_t)(cp + off), (caddr_t)&ipaddr.sin_addr, |
bcopy((caddr_t)(cp + off), (caddr_t)&ipaddr.sin_addr, |
sizeof(ipaddr.sin_addr)); |
sizeof(ipaddr.sin_addr)); |
if (opt == IPOPT_SSRR) { |
if (opt == IPOPT_SSRR) |
#define INA struct in_ifaddr * |
ia = ifatoia(ifa_ifwithaddr(sintosa(&ipaddr))); |
#define SA struct sockaddr * |
else |
ia = (INA)ifa_ifwithladdr((SA)&ipaddr); |
|
} else |
|
ia = ip_rtaddr(ipaddr.sin_addr); |
ia = ip_rtaddr(ipaddr.sin_addr); |
if (ia == 0) { |
if (ia == 0) { |
type = ICMP_UNREACH; |
type = ICMP_UNREACH; |
|
Line 1181 ip_dooptions(m) |
|
break; |
break; |
|
|
case IPOPT_RR: |
case IPOPT_RR: |
|
if (optlen < IPOPT_OFFSET + sizeof(*cp)) { |
|
code = &cp[IPOPT_OLEN] - (u_char *)ip; |
|
goto bad; |
|
} |
if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { |
if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) { |
code = &cp[IPOPT_OFFSET] - (u_char *)ip; |
code = &cp[IPOPT_OFFSET] - (u_char *)ip; |
goto bad; |
goto bad; |
|
Line 1193 ip_dooptions(m) |
|
* If no space remains, ignore. |
* If no space remains, ignore. |
*/ |
*/ |
off--; /* 0 origin */ |
off--; /* 0 origin */ |
if (off > optlen - sizeof(struct in_addr)) |
if ((off + sizeof(struct in_addr)) > optlen) |
break; |
break; |
bcopy((caddr_t)(&ip->ip_dst), (caddr_t)&ipaddr.sin_addr, |
bcopy((caddr_t)(&ip->ip_dst), (caddr_t)&ipaddr.sin_addr, |
sizeof(ipaddr.sin_addr)); |
sizeof(ipaddr.sin_addr)); |
|
Line 1201 ip_dooptions(m) |
|
* locate outgoing interface; if we're the destination, |
* locate outgoing interface; if we're the destination, |
* use the incoming interface (should be same). |
* use the incoming interface (should be same). |
*/ |
*/ |
if ((ia = (INA)ifa_ifwithaddr((SA)&ipaddr)) == 0 && |
if ((ia = ifatoia(ifa_ifwithaddr(sintosa(&ipaddr)))) |
(ia = ip_rtaddr(ipaddr.sin_addr)) == 0) { |
== NULL && |
|
(ia = ip_rtaddr(ipaddr.sin_addr)) == NULL) { |
type = ICMP_UNREACH; |
type = ICMP_UNREACH; |
code = ICMP_UNREACH_HOST; |
code = ICMP_UNREACH_HOST; |
goto bad; |
goto bad; |
|
Line 1216 ip_dooptions(m) |
|
case IPOPT_TS: |
case IPOPT_TS: |
code = cp - (u_char *)ip; |
code = cp - (u_char *)ip; |
ipt = (struct ip_timestamp *)cp; |
ipt = (struct ip_timestamp *)cp; |
if (ipt->ipt_len < 5) |
if (ipt->ipt_len < 4 || ipt->ipt_len > 40) { |
|
code = (u_char *)&ipt->ipt_len - (u_char *)ip; |
|
goto bad; |
|
} |
|
if (ipt->ipt_ptr < 5) { |
|
code = (u_char *)&ipt->ipt_ptr - (u_char *)ip; |
goto bad; |
goto bad; |
|
} |
if (ipt->ipt_ptr > ipt->ipt_len - sizeof (int32_t)) { |
if (ipt->ipt_ptr > ipt->ipt_len - sizeof (int32_t)) { |
if (++ipt->ipt_oflw == 0) |
if (++ipt->ipt_oflw == 0) { |
|
code = (u_char *)&ipt->ipt_ptr - |
|
(u_char *)ip; |
goto bad; |
goto bad; |
|
} |
break; |
break; |
} |
} |
sin = (struct in_addr *)(cp + ipt->ipt_ptr - 1); |
cp0 = (cp + ipt->ipt_ptr - 1); |
switch (ipt->ipt_flg) { |
switch (ipt->ipt_flg) { |
|
|
case IPOPT_TS_TSONLY: |
case IPOPT_TS_TSONLY: |
break; |
break; |
|
|
case IPOPT_TS_TSANDADDR: |
case IPOPT_TS_TSANDADDR: |
if (ipt->ipt_ptr + sizeof(n_time) + |
if (ipt->ipt_ptr - 1 + sizeof(n_time) + |
sizeof(struct in_addr) > ipt->ipt_len) |
sizeof(struct in_addr) > ipt->ipt_len) { |
|
code = (u_char *)&ipt->ipt_ptr - |
|
(u_char *)ip; |
goto bad; |
goto bad; |
|
} |
ipaddr.sin_addr = dst; |
ipaddr.sin_addr = dst; |
ia = (INA)ifaof_ifpforaddr((SA)&ipaddr, |
ia = ifatoia(ifaof_ifpforaddr(sintosa(&ipaddr), |
m->m_pkthdr.rcvif); |
m->m_pkthdr.rcvif)); |
if (ia == 0) |
if (ia == 0) |
continue; |
continue; |
bcopy((caddr_t)&ia->ia_addr.sin_addr, |
bcopy(&ia->ia_addr.sin_addr, |
(caddr_t)sin, sizeof(struct in_addr)); |
cp0, sizeof(struct in_addr)); |
ipt->ipt_ptr += sizeof(struct in_addr); |
ipt->ipt_ptr += sizeof(struct in_addr); |
break; |
break; |
|
|
case IPOPT_TS_PRESPEC: |
case IPOPT_TS_PRESPEC: |
if (ipt->ipt_ptr + sizeof(n_time) + |
if (ipt->ipt_ptr - 1 + sizeof(n_time) + |
sizeof(struct in_addr) > ipt->ipt_len) |
sizeof(struct in_addr) > ipt->ipt_len) { |
|
code = (u_char *)&ipt->ipt_ptr - |
|
(u_char *)ip; |
goto bad; |
goto bad; |
bcopy((caddr_t)sin, (caddr_t)&ipaddr.sin_addr, |
} |
|
bcopy(cp0, &ipaddr.sin_addr, |
sizeof(struct in_addr)); |
sizeof(struct in_addr)); |
if (ifa_ifwithaddr((SA)&ipaddr) == 0) |
if (ifatoia(ifa_ifwithaddr(sintosa(&ipaddr))) |
|
== NULL) |
continue; |
continue; |
ipt->ipt_ptr += sizeof(struct in_addr); |
ipt->ipt_ptr += sizeof(struct in_addr); |
break; |
break; |
|
|
default: |
default: |
|
/* XXX can't take &ipt->ipt_flg */ |
|
code = (u_char *)&ipt->ipt_ptr - |
|
(u_char *)ip + 1; |
goto bad; |
goto bad; |
} |
} |
ntime = iptime(); |
ntime = iptime(); |
bcopy((caddr_t)&ntime, (caddr_t)cp + ipt->ipt_ptr - 1, |
cp0 = (u_char *) &ntime; /* XXX grumble, GCC... */ |
|
bcopy(cp0, (caddr_t)cp + ipt->ipt_ptr - 1, |
sizeof(n_time)); |
sizeof(n_time)); |
ipt->ipt_ptr += sizeof(n_time); |
ipt->ipt_ptr += sizeof(n_time); |
} |
} |
|
Line 1294 ip_dooptions(m) |
|
} |
} |
return (0); |
return (0); |
bad: |
bad: |
ip->ip_len -= ip->ip_hl << 2; /* XXX icmp_error adds in hdr length */ |
|
icmp_error(m, type, code, 0, 0); |
icmp_error(m, type, code, 0, 0); |
ipstat.ips_badoptions++; |
ipstat.ips_badoptions++; |
return (1); |
return (1); |
Line 949 struct in_ifaddr * |
|
Line 1307 struct in_ifaddr * |
|
ip_rtaddr(dst) |
ip_rtaddr(dst) |
struct in_addr dst; |
struct in_addr dst; |
{ |
{ |
register struct sockaddr_in *sin; |
struct sockaddr_in *sin; |
|
|
sin = satosin(&ipforward_rt.ro_dst); |
sin = satosin(&ipforward_rt.ro_dst); |
|
|
Line 984 save_rte(option, dst) |
|
Line 1342 save_rte(option, dst) |
|
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if (ipprintfs) |
if (ipprintfs) |
printf("save_rte: olen %d\n", olen); |
printf("save_rte: olen %d\n", olen); |
#endif |
#endif /* 0 */ |
if (olen > sizeof(ip_srcrt) - (1 + sizeof(dst))) |
if (olen > sizeof(ip_srcrt) - (1 + sizeof(dst))) |
return; |
return; |
bcopy((caddr_t)option, (caddr_t)ip_srcrt.srcopt, olen); |
bcopy((caddr_t)option, (caddr_t)ip_srcrt.srcopt, olen); |
Line 1000 save_rte(option, dst) |
|
Line 1358 save_rte(option, dst) |
|
struct mbuf * |
struct mbuf * |
ip_srcroute() |
ip_srcroute() |
{ |
{ |
register struct in_addr *p, *q; |
struct in_addr *p, *q; |
register struct mbuf *m; |
struct mbuf *m; |
|
|
if (ip_nhops == 0) |
if (ip_nhops == 0) |
return ((struct mbuf *)0); |
return ((struct mbuf *)0); |
|
|
*/ |
*/ |
void |
void |
ip_stripoptions(m, mopt) |
ip_stripoptions(m, mopt) |
register struct mbuf *m; |
struct mbuf *m; |
struct mbuf *mopt; |
struct mbuf *mopt; |
{ |
{ |
register int i; |
int i; |
struct ip *ip = mtod(m, struct ip *); |
struct ip *ip = mtod(m, struct ip *); |
register caddr_t opts; |
caddr_t opts; |
int olen; |
int olen; |
|
|
olen = (ip->ip_hl<<2) - sizeof (struct ip); |
olen = (ip->ip_hl << 2) - sizeof (struct ip); |
opts = (caddr_t)(ip + 1); |
opts = (caddr_t)(ip + 1); |
i = m->m_len - (sizeof (struct ip) + olen); |
i = m->m_len - (sizeof (struct ip) + olen); |
bcopy(opts + olen, opts, (unsigned)i); |
bcopy(opts + olen, opts, (unsigned)i); |
m->m_len -= olen; |
m->m_len -= olen; |
if (m->m_flags & M_PKTHDR) |
if (m->m_flags & M_PKTHDR) |
m->m_pkthdr.len -= olen; |
m->m_pkthdr.len -= olen; |
ip->ip_hl = sizeof(struct ip) >> 2; |
ip->ip_len -= olen; |
|
ip->ip_hl = sizeof (struct ip) >> 2; |
} |
} |
|
|
int inetctlerrmap[PRC_NCMDS] = { |
const int inetctlerrmap[PRC_NCMDS] = { |
0, 0, 0, 0, |
0, 0, 0, 0, |
0, EMSGSIZE, EHOSTDOWN, EHOSTUNREACH, |
0, EMSGSIZE, EHOSTDOWN, EHOSTUNREACH, |
EHOSTUNREACH, EHOSTUNREACH, ECONNREFUSED, ECONNREFUSED, |
EHOSTUNREACH, EHOSTUNREACH, ECONNREFUSED, ECONNREFUSED, |
Line 1116 ip_forward(m, srcrt) |
|
Line 1475 ip_forward(m, srcrt) |
|
struct mbuf *m; |
struct mbuf *m; |
int srcrt; |
int srcrt; |
{ |
{ |
register struct ip *ip = mtod(m, struct ip *); |
struct ip *ip = mtod(m, struct ip *); |
register struct sockaddr_in *sin; |
struct sockaddr_in *sin; |
register struct rtentry *rt; |
struct rtentry *rt; |
int error, type = 0, code = 0; |
int error, type = 0, code = 0; |
struct mbuf *mcopy; |
struct mbuf *mcopy; |
n_long dest; |
n_long dest; |
struct ifnet *destifp; |
struct ifnet *destifp; |
|
#ifdef IPSEC |
|
struct ifnet dummyifp; |
|
#endif |
|
|
|
/* |
|
* Clear any in-bound checksum flags for this packet. |
|
*/ |
|
m->m_pkthdr.csum_flags = 0; |
|
|
dest = 0; |
dest = 0; |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if (ipprintfs) |
if (ipprintfs) |
printf("forward: src %x dst %x ttl %x\n", |
printf("forward: src %2.2x dst %2.2x ttl %x\n", |
ip->ip_src.s_addr, ip->ip_dst.s_addr, ip->ip_ttl); |
ntohl(ip->ip_src.s_addr), |
|
ntohl(ip->ip_dst.s_addr), ip->ip_ttl); |
#endif |
#endif |
if (m->m_flags & M_BCAST || in_canforward(ip->ip_dst) == 0) { |
if (m->m_flags & (M_BCAST|M_MCAST) || in_canforward(ip->ip_dst) == 0) { |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
m_freem(m); |
m_freem(m); |
return; |
return; |
} |
} |
HTONS(ip->ip_id); |
|
if (ip->ip_ttl <= IPTTLDEC) { |
if (ip->ip_ttl <= IPTTLDEC) { |
icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, dest, 0); |
icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, dest, 0); |
return; |
return; |
Line 1164 ip_forward(m, srcrt) |
|
Line 1531 ip_forward(m, srcrt) |
|
/* |
/* |
* Save at most 68 bytes of the packet in case |
* Save at most 68 bytes of the packet in case |
* we need to generate an ICMP message to the src. |
* we need to generate an ICMP message to the src. |
|
* Pullup to avoid sharing mbuf cluster between m and mcopy. |
*/ |
*/ |
mcopy = m_copy(m, 0, imin((int)ip->ip_len, 68)); |
mcopy = m_copym(m, 0, imin((int)ip->ip_len, 68), M_DONTWAIT); |
|
if (mcopy) |
|
mcopy = m_pullup(mcopy, ip->ip_hl << 2); |
|
|
/* |
/* |
* If forwarding packet using same interface that it came in on, |
* If forwarding packet using same interface that it came in on, |
Line 1182 ip_forward(m, srcrt) |
|
Line 1552 ip_forward(m, srcrt) |
|
if (rt->rt_ifa && |
if (rt->rt_ifa && |
(ip->ip_src.s_addr & ifatoia(rt->rt_ifa)->ia_subnetmask) == |
(ip->ip_src.s_addr & ifatoia(rt->rt_ifa)->ia_subnetmask) == |
ifatoia(rt->rt_ifa)->ia_subnet) { |
ifatoia(rt->rt_ifa)->ia_subnet) { |
if (rt->rt_flags & RTF_GATEWAY) |
if (rt->rt_flags & RTF_GATEWAY) |
dest = satosin(rt->rt_gateway)->sin_addr.s_addr; |
dest = satosin(rt->rt_gateway)->sin_addr.s_addr; |
else |
else |
dest = ip->ip_dst.s_addr; |
dest = ip->ip_dst.s_addr; |
/* Router requirements says to only send host redirects */ |
/* |
type = ICMP_REDIRECT; |
* Router requirements says to only send host |
code = ICMP_REDIRECT_HOST; |
* redirects. |
|
*/ |
|
type = ICMP_REDIRECT; |
|
code = ICMP_REDIRECT_HOST; |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if (ipprintfs) |
if (ipprintfs) |
printf("redirect (%d) to %x\n", code, (u_int32_t)dest); |
printf("redirect (%d) to %x\n", code, |
|
(u_int32_t)dest); |
#endif |
#endif |
} |
} |
} |
} |
|
|
|
#ifdef IPSEC |
|
/* Don't lookup socket in forwarding case */ |
|
(void)ipsec_setsocket(m, NULL); |
|
#endif |
error = ip_output(m, (struct mbuf *)0, &ipforward_rt, |
error = ip_output(m, (struct mbuf *)0, &ipforward_rt, |
(IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 0); |
(IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 0); |
if (error) |
if (error) |
Line 1237 ip_forward(m, srcrt) |
|
Line 1615 ip_forward(m, srcrt) |
|
case EMSGSIZE: |
case EMSGSIZE: |
type = ICMP_UNREACH; |
type = ICMP_UNREACH; |
code = ICMP_UNREACH_NEEDFRAG; |
code = ICMP_UNREACH_NEEDFRAG; |
|
#ifndef IPSEC |
if (ipforward_rt.ro_rt) |
if (ipforward_rt.ro_rt) |
destifp = ipforward_rt.ro_rt->rt_ifp; |
destifp = ipforward_rt.ro_rt->rt_ifp; |
|
#else |
|
/* |
|
* If the packet is routed over IPsec tunnel, tell the |
|
* originator the tunnel MTU. |
|
* tunnel MTU = if MTU - sizeof(IP) - ESP/AH hdrsiz |
|
* XXX quickhack!!! |
|
*/ |
|
if (ipforward_rt.ro_rt) { |
|
struct secpolicy *sp; |
|
int ipsecerror; |
|
size_t ipsechdr; |
|
struct route *ro; |
|
|
|
sp = ipsec4_getpolicybyaddr(mcopy, |
|
IPSEC_DIR_OUTBOUND, |
|
IP_FORWARDING, |
|
&ipsecerror); |
|
|
|
if (sp == NULL) |
|
destifp = ipforward_rt.ro_rt->rt_ifp; |
|
else { |
|
/* count IPsec header size */ |
|
ipsechdr = ipsec4_hdrsiz(mcopy, |
|
IPSEC_DIR_OUTBOUND, |
|
NULL); |
|
|
|
/* |
|
* find the correct route for outer IPv4 |
|
* header, compute tunnel MTU. |
|
* |
|
* XXX BUG ALERT |
|
* The "dummyifp" code relies upon the fact |
|
* that icmp_error() touches only ifp->if_mtu. |
|
*/ |
|
/*XXX*/ |
|
destifp = NULL; |
|
if (sp->req != NULL |
|
&& sp->req->sav != NULL |
|
&& sp->req->sav->sah != NULL) { |
|
ro = &sp->req->sav->sah->sa_route; |
|
if (ro->ro_rt && ro->ro_rt->rt_ifp) { |
|
dummyifp.if_mtu = |
|
ro->ro_rt->rt_ifp->if_mtu; |
|
dummyifp.if_mtu -= ipsechdr; |
|
destifp = &dummyifp; |
|
} |
|
} |
|
|
|
key_freesp(sp); |
|
} |
|
} |
|
#endif /*IPSEC*/ |
ipstat.ips_cantfrag++; |
ipstat.ips_cantfrag++; |
break; |
break; |
|
|
case ENOBUFS: |
case ENOBUFS: |
|
#if 1 |
|
/* |
|
* a router should not generate ICMP_SOURCEQUENCH as |
|
* required in RFC1812 Requirements for IP Version 4 Routers. |
|
* source quench could be a big problem under DoS attacks, |
|
* or the underlying interface is rate-limited. |
|
*/ |
|
if (mcopy) |
|
m_freem(mcopy); |
|
return; |
|
#else |
type = ICMP_SOURCEQUENCH; |
type = ICMP_SOURCEQUENCH; |
code = 0; |
code = 0; |
break; |
break; |
|
#endif |
} |
} |
icmp_error(mcopy, type, code, dest, destifp); |
icmp_error(mcopy, type, code, dest, destifp); |
} |
} |
|
|
void |
void |
ip_savecontrol(inp, mp, ip, m) |
ip_savecontrol(inp, mp, ip, m) |
register struct inpcb *inp; |
struct inpcb *inp; |
register struct mbuf **mp; |
struct mbuf **mp; |
register struct ip *ip; |
struct ip *ip; |
register struct mbuf *m; |
struct mbuf *m; |
{ |
{ |
|
|
if (inp->inp_socket->so_options & SO_TIMESTAMP) { |
if (inp->inp_socket->so_options & SO_TIMESTAMP) { |
Line 1319 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1762 ip_sysctl(name, namelen, oldp, oldlenp, |
|
void *newp; |
void *newp; |
size_t newlen; |
size_t newlen; |
{ |
{ |
extern int subnetsarelocal; |
extern int subnetsarelocal, hostzeroisbroadcast; |
|
|
int error, old; |
int error, old; |
|
|
Line 1370 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1813 ip_sysctl(name, namelen, oldp, oldlenp, |
|
case IPCTL_ANONPORTMIN: |
case IPCTL_ANONPORTMIN: |
old = anonportmin; |
old = anonportmin; |
error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmin); |
error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmin); |
if (anonportmin >= anonportmax || anonportmin > 65535 |
if (anonportmin >= anonportmax || anonportmin < 0 |
|
|| anonportmin > 65535 |
#ifndef IPNOPRIVPORTS |
#ifndef IPNOPRIVPORTS |
|| anonportmin < IPPORT_RESERVED |
|| anonportmin < IPPORT_RESERVED |
#endif |
#endif |
Line 1382 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1826 ip_sysctl(name, namelen, oldp, oldlenp, |
|
case IPCTL_ANONPORTMAX: |
case IPCTL_ANONPORTMAX: |
old = anonportmax; |
old = anonportmax; |
error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmax); |
error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmax); |
if (anonportmin >= anonportmax || anonportmax > 65535 |
if (anonportmin >= anonportmax || anonportmax < 0 |
|
|| anonportmax > 65535 |
#ifndef IPNOPRIVPORTS |
#ifndef IPNOPRIVPORTS |
|| anonportmax < IPPORT_RESERVED |
|| anonportmax < IPPORT_RESERVED |
#endif |
#endif |
Line 1398 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1843 ip_sysctl(name, namelen, oldp, oldlenp, |
|
rt_timer_queue_change(ip_mtudisc_timeout_q, |
rt_timer_queue_change(ip_mtudisc_timeout_q, |
ip_mtudisc_timeout); |
ip_mtudisc_timeout); |
return (error); |
return (error); |
|
#ifdef GATEWAY |
|
case IPCTL_MAXFLOWS: |
|
{ |
|
int s; |
|
|
|
error = sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_maxflows); |
|
s = splsoftnet(); |
|
ipflow_reap(0); |
|
splx(s); |
|
return (error); |
|
} |
|
#endif |
|
case IPCTL_HOSTZEROBROADCAST: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&hostzeroisbroadcast)); |
|
#if NGIF > 0 |
|
case IPCTL_GIF_TTL: |
|
return(sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_gif_ttl)); |
|
#endif |
|
|
|
#if NGRE > 0 |
|
case IPCTL_GRE_TTL: |
|
return(sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_gre_ttl)); |
|
#endif |
|
|
|
#ifndef IPNOPRIVPORTS |
|
case IPCTL_LOWPORTMIN: |
|
old = lowportmin; |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, &lowportmin); |
|
if (lowportmin >= lowportmax |
|
|| lowportmin > IPPORT_RESERVEDMAX |
|
|| lowportmin < IPPORT_RESERVEDMIN |
|
) { |
|
lowportmin = old; |
|
return (EINVAL); |
|
} |
|
return (error); |
|
case IPCTL_LOWPORTMAX: |
|
old = lowportmax; |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, &lowportmax); |
|
if (lowportmin >= lowportmax |
|
|| lowportmax > IPPORT_RESERVEDMAX |
|
|| lowportmax < IPPORT_RESERVEDMIN |
|
) { |
|
lowportmax = old; |
|
return (EINVAL); |
|
} |
|
return (error); |
|
#endif |
|
|
|
case IPCTL_MAXFRAGPACKETS: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_maxfragpackets)); |
|
|
default: |
default: |
return (EOPNOTSUPP); |
return (EOPNOTSUPP); |
} |
} |