version 1.114.4.5, 2001/04/06 00:24:47 |
version 1.179, 2003/09/28 04:45:14 |
|
|
/* |
/* |
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. |
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. |
* All rights reserved. |
* All rights reserved. |
* |
* |
* Redistribution and use in source and binary forms, with or without |
* Redistribution and use in source and binary forms, with or without |
* modification, are permitted provided that the following conditions |
* modification, are permitted provided that the following conditions |
* are met: |
* are met: |
|
|
* 3. Neither the name of the project nor the names of its contributors |
* 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 |
* may be used to endorse or promote products derived from this software |
* without specific prior written permission. |
* without specific prior written permission. |
* |
* |
* THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND |
* THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND |
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
|
|
* 2. Redistributions in binary form must reproduce the above copyright |
* 2. Redistributions in binary form must reproduce the above copyright |
* notice, this list of conditions and the following disclaimer in the |
* notice, this list of conditions and the following disclaimer in the |
* documentation and/or other materials provided with the distribution. |
* documentation and/or other materials provided with the distribution. |
* 3. All advertising materials mentioning features or use of this software |
* 3. Neither the name of the University nor the names of its contributors |
* 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 |
* may be used to endorse or promote products derived from this software |
* without specific prior written permission. |
* without specific prior written permission. |
* |
* |
|
|
* @(#)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_pfil_hooks.h" |
#include "opt_ipsec.h" |
#include "opt_ipsec.h" |
#include "opt_mrouting.h" |
#include "opt_mrouting.h" |
|
#include "opt_mbuftrace.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 <sys/pool.h> |
|
|
#include <vm/vm.h> |
|
#include <sys/sysctl.h> |
#include <sys/sysctl.h> |
|
|
#include <net/if.h> |
#include <net/if.h> |
|
|
/* just for gif_ttl */ |
/* just for gif_ttl */ |
#include <netinet/in_gif.h> |
#include <netinet/in_gif.h> |
#include "gif.h" |
#include "gif.h" |
|
#include <net/if_gre.h> |
|
#include "gre.h" |
|
|
#ifdef MROUTING |
#ifdef MROUTING |
#include <netinet/ip_mroute.h> |
#include <netinet/ip_mroute.h> |
|
|
#ifdef IPSEC |
#ifdef IPSEC |
#include <netinet6/ipsec.h> |
#include <netinet6/ipsec.h> |
#include <netkey/key.h> |
#include <netkey/key.h> |
#include <netkey/key_debug.h> |
|
#endif |
#endif |
|
#ifdef FAST_IPSEC |
|
#include <netipsec/ipsec.h> |
|
#include <netipsec/key.h> |
|
#endif /* FAST_IPSEC*/ |
|
|
#ifndef IPFORWARDING |
#ifndef IPFORWARDING |
#ifdef GATEWAY |
#ifdef GATEWAY |
|
|
#define IPALLOWSRCRT 1 /* allow source-routed packets */ |
#define IPALLOWSRCRT 1 /* allow source-routed packets */ |
#endif |
#endif |
#ifndef IPMTUDISC |
#ifndef IPMTUDISC |
#define IPMTUDISC 0 |
#define IPMTUDISC 1 |
#endif |
#endif |
#ifndef IPMTUDISCTIMEOUT |
#ifndef IPMTUDISCTIMEOUT |
#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ |
#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ |
Line 190 int ip_forwsrcrt = IPFORWSRCRT; |
|
Line 193 int ip_forwsrcrt = IPFORWSRCRT; |
|
int ip_directedbcast = IPDIRECTEDBCAST; |
int ip_directedbcast = IPDIRECTEDBCAST; |
int ip_allowsrcrt = IPALLOWSRCRT; |
int ip_allowsrcrt = IPALLOWSRCRT; |
int ip_mtudisc = IPMTUDISC; |
int ip_mtudisc = IPMTUDISC; |
u_int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; |
int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
int ipprintfs = 0; |
int ipprintfs = 0; |
#endif |
#endif |
|
/* |
|
* XXX - Setting ip_checkinterface mostly implements the receive side of |
|
* the Strong ES model described in RFC 1122, but since the routing table |
|
* and transmit implementation do not implement the Strong ES model, |
|
* setting this to 1 results in an odd hybrid. |
|
* |
|
* XXX - ip_checkinterface currently must be disabled if you use ipnat |
|
* to translate the destination address to another local interface. |
|
* |
|
* XXX - ip_checkinterface must be disabled if you add IP aliases |
|
* to the loopback interface instead of the interface where the |
|
* packets for those addresses are received. |
|
*/ |
|
int ip_checkinterface = 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; |
int ipqmaxlen = IFQ_MAXLEN; |
int ipqmaxlen = IFQ_MAXLEN; |
|
u_long in_ifaddrhash; /* size of hash table - 1 */ |
|
int in_ifaddrentries; /* total number of addrs */ |
struct in_ifaddrhead in_ifaddr; |
struct in_ifaddrhead in_ifaddr; |
struct in_ifaddrhashhead *in_ifaddrhashtbl; |
struct in_ifaddrhashhead *in_ifaddrhashtbl; |
|
u_long in_multihash; /* size of hash table - 1 */ |
|
int in_multientries; /* total number of addrs */ |
|
struct in_multihashhead *in_multihashtbl; |
struct ifqueue ipintrq; |
struct ifqueue ipintrq; |
struct ipstat ipstat; |
struct ipstat ipstat; |
u_int16_t ip_id; |
|
|
#ifdef PFIL_HOOKS |
|
struct pfil_head inet_pfil_hook; |
|
#endif |
|
|
struct ipqhead ipq; |
struct ipqhead ipq; |
int ipq_locked; |
int ipq_locked; |
|
int ip_nfragpackets = 0; |
|
int ip_maxfragpackets = 200; |
|
|
static __inline int ipq_lock_try __P((void)); |
static __inline int ipq_lock_try __P((void)); |
static __inline void ipq_unlock __P((void)); |
static __inline void ipq_unlock __P((void)); |
|
|
{ |
{ |
int s; |
int s; |
|
|
s = splimp(); |
/* |
|
* Use splvm() -- we're blocking things that would cause |
|
* mbuf allocation. |
|
*/ |
|
s = splvm(); |
if (ipq_locked) { |
if (ipq_locked) { |
splx(s); |
splx(s); |
return (0); |
return (0); |
|
|
{ |
{ |
int s; |
int s; |
|
|
s = splimp(); |
s = splvm(); |
ipq_locked = 0; |
ipq_locked = 0; |
splx(s); |
splx(s); |
} |
} |
|
|
printf("%s:%d: ipq already locked\n", __FILE__, __LINE__); \ |
printf("%s:%d: ipq already locked\n", __FILE__, __LINE__); \ |
panic("ipq_lock"); \ |
panic("ipq_lock"); \ |
} \ |
} \ |
} while (0) |
} while (/*CONSTCOND*/ 0) |
#define IPQ_LOCK_CHECK() \ |
#define IPQ_LOCK_CHECK() \ |
do { \ |
do { \ |
if (ipq_locked == 0) { \ |
if (ipq_locked == 0) { \ |
printf("%s:%d: ipq lock not held\n", __FILE__, __LINE__); \ |
printf("%s:%d: ipq lock not held\n", __FILE__, __LINE__); \ |
panic("ipq lock check"); \ |
panic("ipq lock check"); \ |
} \ |
} \ |
} while (0) |
} while (/*CONSTCOND*/ 0) |
#else |
#else |
#define IPQ_LOCK() (void) ipq_lock_try() |
#define IPQ_LOCK() (void) ipq_lock_try() |
#define IPQ_LOCK_CHECK() /* nothing */ |
#define IPQ_LOCK_CHECK() /* nothing */ |
|
|
|
|
#define IPQ_UNLOCK() ipq_unlock() |
#define IPQ_UNLOCK() ipq_unlock() |
|
|
|
struct pool inmulti_pool; |
struct pool ipqent_pool; |
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 |
* to an incoming packet over the same route if the packet got here |
* to an incoming packet over the same route if the packet got here |
Line 277 static struct ip_srcrt { |
|
Line 328 static struct ip_srcrt { |
|
|
|
static void save_rte __P((u_char *, struct in_addr)); |
static void save_rte __P((u_char *, struct in_addr)); |
|
|
|
#ifdef MBUFTRACE |
|
struct mowner ip_rx_mowner = { "internet", "rx" }; |
|
struct mowner ip_tx_mowner = { "internet", "tx" }; |
|
#endif |
|
|
/* |
/* |
* IP initialization: fill in IP protocol switch table. |
* IP initialization: fill in IP protocol switch table. |
* All protocols not implemented in kernel go to raw IP protocol handler. |
* All protocols not implemented in kernel go to raw IP protocol handler. |
|
|
struct protosw *pr; |
struct protosw *pr; |
int i; |
int i; |
|
|
|
pool_init(&inmulti_pool, sizeof(struct in_multi), 0, 0, 0, "inmltpl", |
|
NULL); |
pool_init(&ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", |
pool_init(&ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", |
0, NULL, NULL, M_IPQ); |
NULL); |
|
|
pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); |
pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); |
if (pr == 0) |
if (pr == 0) |
|
|
pr->pr_protocol && pr->pr_protocol != IPPROTO_RAW) |
pr->pr_protocol && pr->pr_protocol != IPPROTO_RAW) |
ip_protox[pr->pr_protocol] = pr - inetsw; |
ip_protox[pr->pr_protocol] = pr - inetsw; |
LIST_INIT(&ipq); |
LIST_INIT(&ipq); |
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) |
in_multihashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IPMADDR, |
ip_mtudisc_timeout_q = |
M_WAITOK, &in_multihash); |
rt_timer_queue_create(ip_mtudisc_timeout); |
ip_mtudisc_timeout_q = rt_timer_queue_create(ip_mtudisc_timeout); |
#ifdef GATEWAY |
#ifdef GATEWAY |
ipflow_init(); |
ipflow_init(); |
#endif |
#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 */ |
|
|
|
#ifdef MBUFTRACE |
|
MOWNER_ATTACH(&ip_tx_mowner); |
|
MOWNER_ATTACH(&ip_rx_mowner); |
|
#endif /* MBUFTRACE */ |
} |
} |
|
|
struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; |
struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET }; |
|
|
struct mbuf *m; |
struct mbuf *m; |
|
|
while (1) { |
while (1) { |
s = splimp(); |
s = splnet(); |
IF_DEQUEUE(&ipintrq, m); |
IF_DEQUEUE(&ipintrq, m); |
splx(s); |
splx(s); |
if (m == 0) |
if (m == 0) |
return; |
return; |
|
MCLAIM(m, &ip_rx_mowner); |
ip_input(m); |
ip_input(m); |
} |
} |
} |
} |
Line 350 ip_input(struct mbuf *m) |
|
Line 429 ip_input(struct mbuf *m) |
|
struct ipqent *ipqe; |
struct ipqent *ipqe; |
int hlen = 0, mff, len; |
int hlen = 0, mff, len; |
int downmatch; |
int downmatch; |
#ifdef PFIL_HOOKS |
int checkif; |
struct packet_filter_hook *pfh; |
int srcrt = 0; |
struct mbuf *m0; |
#ifdef FAST_IPSEC |
int rv; |
struct m_tag *mtag; |
#endif /* PFIL_HOOKS */ |
struct tdb_ident *tdbi; |
|
struct secpolicy *sp; |
|
int s, error; |
|
#endif /* FAST_IPSEC */ |
|
|
|
MCLAIM(m, &ip_rx_mowner); |
#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) && |
/* |
(m = m_pullup(m, sizeof (struct ip))) == 0) { |
* If the IP header is not aligned, slurp it up into a new |
ipstat.ips_toosmall++; |
* mbuf with space for link headers, in the event we forward |
return; |
* it. Otherwise, if it is aligned, make sure the entire |
|
* base IP header is in the first mbuf of the chain. |
|
*/ |
|
if (IP_HDR_ALIGNED_P(mtod(m, caddr_t)) == 0) { |
|
if ((m = m_copyup(m, sizeof(struct ip), |
|
(max_linkhdr + 3) & ~3)) == NULL) { |
|
/* XXXJRT new stat, please */ |
|
ipstat.ips_toosmall++; |
|
return; |
|
} |
|
} else if (__predict_false(m->m_len < sizeof (struct ip))) { |
|
if ((m = m_pullup(m, sizeof (struct ip))) == NULL) { |
|
ipstat.ips_toosmall++; |
|
return; |
|
} |
} |
} |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
if (ip->ip_v != IPVERSION) { |
if (ip->ip_v != IPVERSION) { |
Line 405 ip_input(struct mbuf *m) |
|
Line 493 ip_input(struct mbuf *m) |
|
* not allowed. |
* not allowed. |
*/ |
*/ |
if (IN_MULTICAST(ip->ip_src.s_addr)) { |
if (IN_MULTICAST(ip->ip_src.s_addr)) { |
/* XXX stat */ |
ipstat.ips_badaddr++; |
goto bad; |
goto bad; |
} |
} |
|
|
if (in_cksum(m, hlen) != 0) { |
/* 127/8 must not appear on wire - RFC1122 */ |
ipstat.ips_badsum++; |
if ((ntohl(ip->ip_dst.s_addr) >> IN_CLASSA_NSHIFT) == IN_LOOPBACKNET || |
goto bad; |
(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 & |
* Convert fields to host representation. |
((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_IPv4) | |
*/ |
M_CSUM_IPv4_BAD)) { |
NTOHS(ip->ip_len); |
case M_CSUM_IPv4|M_CSUM_IPv4_BAD: |
NTOHS(ip->ip_off); |
INET_CSUM_COUNTER_INCR(&ip_hwcsum_bad); |
len = ip->ip_len; |
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); |
|
|
/* |
/* |
* Check for additional length bogosity |
* Check for additional length bogosity |
Line 448 ip_input(struct mbuf *m) |
|
Line 556 ip_input(struct mbuf *m) |
|
} |
} |
|
|
#ifdef IPSEC |
#ifdef IPSEC |
/* ipflow (IP fast fowarding) is not compatible with IPsec. */ |
/* ipflow (IP fast forwarding) is not compatible with IPsec. */ |
m->m_flags &= ~M_CANFASTFWD; |
m->m_flags &= ~M_CANFASTFWD; |
#else |
#else |
/* |
/* |
Line 459 ip_input(struct mbuf *m) |
|
Line 567 ip_input(struct mbuf *m) |
|
#endif |
#endif |
|
|
#ifdef PFIL_HOOKS |
#ifdef PFIL_HOOKS |
#ifdef IPSEC |
|
/* |
|
* let ipfilter look at packet on the wire, |
|
* not the decapsulated packet. |
|
*/ |
|
if (ipsec_gethist(m, NULL)) |
|
goto nofilt; |
|
#endif |
|
/* |
/* |
* Run through list of hooks for input packets. If there are any |
* Run through list of hooks for input packets. If there are any |
* filters which require that additional packets in the flow are |
* filters which require that additional packets in the flow are |
Line 474 ip_input(struct mbuf *m) |
|
Line 574 ip_input(struct mbuf *m) |
|
* 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; |
/* |
pfh = pfil_hook_get(PFIL_IN, &inetsw[ip_protox[IPPROTO_IP]].pr_pfh); |
* let ipfilter look at packet on the wire, |
for (; pfh; pfh = pfh->pfil_link.tqe_next) |
* not the decapsulated packet. |
if (pfh->pfil_func) { |
*/ |
rv = pfh->pfil_func(ip, hlen, |
|
m->m_pkthdr.rcvif, 0, &m0); |
|
if (rv) |
|
return; |
|
m = m0; |
|
if (m == NULL) |
|
return; |
|
ip = mtod(m, struct ip *); |
|
} |
|
#ifdef IPSEC |
#ifdef IPSEC |
nofilt:; |
if (!ipsec_getnhist(m)) |
|
#else |
|
if (1) |
#endif |
#endif |
|
{ |
|
struct in_addr odst; |
|
|
|
odst = ip->ip_dst; |
|
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; |
|
srcrt = (odst.s_addr != ip->ip_dst.s_addr); |
|
} |
#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 |
|
|
/* |
/* |
* 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 |
|
|
return; |
return; |
|
|
/* |
/* |
|
* Enable a consistency check between the destination address |
|
* and the arrival interface for a unicast packet (the RFC 1122 |
|
* strong ES model) if IP forwarding is disabled and the packet |
|
* is not locally generated. |
|
* |
|
* XXX - Checking also should be disabled if the destination |
|
* address is ipnat'ed to a different interface. |
|
* |
|
* XXX - Checking is incompatible with IP aliases added |
|
* to the loopback interface instead of the interface where |
|
* the packets are received. |
|
* |
|
* XXX - We need to add a per ifaddr flag for this so that |
|
* we get finer grain control. |
|
*/ |
|
checkif = ip_checkinterface && (ipforwarding == 0) && |
|
(m->m_pkthdr.rcvif != NULL) && |
|
((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0); |
|
|
|
/* |
* 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. |
* |
* |
* Traditional 4.4BSD did not consult IFF_UP at all. |
* Traditional 4.4BSD did not consult IFF_UP at all. |
|
|
* as not mine. |
* as not mine. |
*/ |
*/ |
downmatch = 0; |
downmatch = 0; |
for (ia = IN_IFADDR_HASH(ip->ip_dst.s_addr).lh_first; |
LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { |
ia != NULL; |
|
ia = ia->ia_hash.le_next) { |
|
if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { |
if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { |
|
if (checkif && ia->ia_ifp != m->m_pkthdr.rcvif) |
|
continue; |
if ((ia->ia_ifp->if_flags & IFF_UP) != 0) |
if ((ia->ia_ifp->if_flags & IFF_UP) != 0) |
break; |
break; |
else |
else |
|
|
if (ia != NULL) |
if (ia != NULL) |
goto ours; |
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++; |
return; |
return; |
|
|
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
return; |
return; |
} |
} |
ip_forward(m, 0); |
#ifdef IPSEC |
|
if (ipsec4_in_reject(m, NULL)) { |
|
ipsecstat.in_polvio++; |
|
goto bad; |
|
} |
|
#endif |
|
#ifdef FAST_IPSEC |
|
mtag = m_tag_find(m, PACKET_TAG_IPSEC_IN_DONE, NULL); |
|
s = splsoftnet(); |
|
if (mtag != NULL) { |
|
tdbi = (struct tdb_ident *)(mtag + 1); |
|
sp = ipsec_getpolicy(tdbi, IPSEC_DIR_INBOUND); |
|
} else { |
|
sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_INBOUND, |
|
IP_FORWARDING, &error); |
|
} |
|
if (sp == NULL) { /* NB: can happen if error */ |
|
splx(s); |
|
/*XXX error stat???*/ |
|
DPRINTF(("ip_input: no SP for forwarding\n")); /*XXX*/ |
|
goto bad; |
|
} |
|
|
|
/* |
|
* Check security policy against packet attributes. |
|
*/ |
|
error = ipsec_in_reject(sp, m); |
|
KEY_FREESP(&sp); |
|
splx(s); |
|
if (error) { |
|
ipstat.ips_cantforward++; |
|
goto bad; |
|
} |
|
#endif /* FAST_IPSEC */ |
|
|
|
ip_forward(m, srcrt); |
} |
} |
return; |
return; |
|
|
|
|
* if the packet was previously fragmented, |
* if the packet was previously fragmented, |
* but it's not worth the time; just let them time out.) |
* but it's not worth the time; just let them time out.) |
*/ |
*/ |
if (ip->ip_off & ~(IP_DF|IP_RF)) { |
if (ip->ip_off & ~htons(IP_DF|IP_RF)) { |
|
if (M_READONLY(m)) { |
|
if ((m = m_pullup(m, hlen)) == NULL) { |
|
ipstat.ips_toosmall++; |
|
goto bad; |
|
} |
|
ip = mtod(m, struct ip *); |
|
} |
|
|
/* |
/* |
* Look for queue of fragments |
* Look for queue of fragments |
* of this datagram. |
* of this datagram. |
*/ |
*/ |
IPQ_LOCK(); |
IPQ_LOCK(); |
for (fp = ipq.lh_first; fp != NULL; fp = fp->ipq_q.le_next) |
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) && |
|
|
* set ipqe_mff if more fragments are expected, |
* set ipqe_mff if more fragments are expected, |
* convert offset of this to bytes. |
* convert offset of this to bytes. |
*/ |
*/ |
ip->ip_len -= hlen; |
ip->ip_len = htons(ntohs(ip->ip_len) - hlen); |
mff = (ip->ip_off & IP_MF) != 0; |
mff = (ip->ip_off & htons(IP_MF)) != 0; |
if (mff) { |
if (mff) { |
/* |
/* |
* Make sure that fragments have a data length |
* Make sure that fragments have a data length |
* that's a non-zero multiple of 8 bytes. |
* that's a non-zero multiple of 8 bytes. |
*/ |
*/ |
if (ip->ip_len == 0 || (ip->ip_len & 0x7) != 0) { |
if (ntohs(ip->ip_len) == 0 || |
|
(ntohs(ip->ip_len) & 0x7) != 0) { |
ipstat.ips_badfrags++; |
ipstat.ips_badfrags++; |
IPQ_UNLOCK(); |
IPQ_UNLOCK(); |
goto bad; |
goto bad; |
} |
} |
} |
} |
ip->ip_off <<= 3; |
ip->ip_off = htons((ntohs(ip->ip_off) & IP_OFFMASK) << 3); |
|
|
/* |
/* |
* If datagram marked as having more fragments |
* If datagram marked as having more fragments |
* or if this is not the first fragment, |
* or if this is not the first fragment, |
* attempt reassembly; if it succeeds, proceed. |
* attempt reassembly; if it succeeds, proceed. |
*/ |
*/ |
if (mff || ip->ip_off) { |
if (mff || ip->ip_off != htons(0)) { |
ipstat.ips_fragments++; |
ipstat.ips_fragments++; |
ipqe = pool_get(&ipqent_pool, PR_NOWAIT); |
ipqe = pool_get(&ipqent_pool, PR_NOWAIT); |
if (ipqe == NULL) { |
if (ipqe == NULL) { |
|
|
ipstat.ips_reassembled++; |
ipstat.ips_reassembled++; |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
hlen = ip->ip_hl << 2; |
hlen = ip->ip_hl << 2; |
ip->ip_len += hlen; |
ip->ip_len = htons(ntohs(ip->ip_len) + hlen); |
} else |
} else |
if (fp) |
if (fp) |
ip_freef(fp); |
ip_freef(fp); |
IPQ_UNLOCK(); |
IPQ_UNLOCK(); |
} |
} |
|
|
#ifdef IPSEC |
#if defined(IPSEC) |
/* |
/* |
* enforce IPsec policy checking if we are seeing last header. |
* enforce IPsec policy checking if we are seeing last header. |
* note that we do not visit this with protocols with pcb layer |
* note that we do not visit this with protocols with pcb layer |
|
|
goto bad; |
goto bad; |
} |
} |
#endif |
#endif |
|
#if FAST_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) { |
|
/* |
|
* Check if the packet has already had IPsec processing |
|
* done. If so, then just pass it along. This tag gets |
|
* set during AH, ESP, etc. input handling, before the |
|
* packet is returned to the ip input queue for delivery. |
|
*/ |
|
mtag = m_tag_find(m, PACKET_TAG_IPSEC_IN_DONE, NULL); |
|
s = splsoftnet(); |
|
if (mtag != NULL) { |
|
tdbi = (struct tdb_ident *)(mtag + 1); |
|
sp = ipsec_getpolicy(tdbi, IPSEC_DIR_INBOUND); |
|
} else { |
|
sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_INBOUND, |
|
IP_FORWARDING, &error); |
|
} |
|
if (sp != NULL) { |
|
/* |
|
* Check security policy against packet attributes. |
|
*/ |
|
error = ipsec_in_reject(sp, m); |
|
KEY_FREESP(&sp); |
|
} else { |
|
/* XXX error stat??? */ |
|
error = EINVAL; |
|
DPRINTF(("ip_input: no SP, packet discarded\n"));/*XXX*/ |
|
goto bad; |
|
} |
|
splx(s); |
|
if (error) |
|
goto bad; |
|
} |
|
#endif /* FAST_IPSEC */ |
|
|
/* |
/* |
* Switch out to protocol's input routine. |
* Switch out to protocol's input routine. |
*/ |
*/ |
#if IFA_STATS |
#if IFA_STATS |
ia->ia_ifa.ifa_data.ifad_inbytes += ip->ip_len; |
if (ia && ip) |
|
ia->ia_ifa.ifa_data.ifad_inbytes += ntohs(ip->ip_len); |
#endif |
#endif |
ipstat.ips_delivered++; |
ipstat.ips_delivered++; |
{ |
{ |
|
|
} |
} |
bad: |
bad: |
m_freem(m); |
m_freem(m); |
|
return; |
|
|
|
badcsum: |
|
ipstat.ips_badsum++; |
|
m_freem(m); |
} |
} |
|
|
/* |
/* |
Line 759 ip_reass(ipqe, fp) |
|
Line 982 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 767 ip_reass(ipqe, fp) |
|
Line 1001 ip_reass(ipqe, fp) |
|
fp->ipq_ttl = IPFRAGTTL; |
fp->ipq_ttl = IPFRAGTTL; |
fp->ipq_p = ipqe->ipqe_ip->ip_p; |
fp->ipq_p = ipqe->ipqe_ip->ip_p; |
fp->ipq_id = ipqe->ipqe_ip->ip_id; |
fp->ipq_id = ipqe->ipqe_ip->ip_id; |
LIST_INIT(&fp->ipq_fragq); |
TAILQ_INIT(&fp->ipq_fragq); |
fp->ipq_src = ipqe->ipqe_ip->ip_src; |
fp->ipq_src = ipqe->ipqe_ip->ip_src; |
fp->ipq_dst = ipqe->ipqe_ip->ip_dst; |
fp->ipq_dst = ipqe->ipqe_ip->ip_dst; |
p = NULL; |
p = NULL; |
Line 777 ip_reass(ipqe, fp) |
|
Line 1011 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 = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = q->ipqe_q.le_next) |
p = q, q = TAILQ_NEXT(q, ipqe_q)) |
if (q->ipqe_ip->ip_off > ipqe->ipqe_ip->ip_off) |
if (ntohs(q->ipqe_ip->ip_off) > ntohs(ipqe->ipqe_ip->ip_off)) |
break; |
break; |
|
|
/* |
/* |
Line 788 ip_reass(ipqe, fp) |
|
Line 1022 ip_reass(ipqe, fp) |
|
* segment. If it provides all of our data, drop us. |
* segment. If it provides all of our data, drop us. |
*/ |
*/ |
if (p != NULL) { |
if (p != NULL) { |
i = p->ipqe_ip->ip_off + p->ipqe_ip->ip_len - |
i = ntohs(p->ipqe_ip->ip_off) + ntohs(p->ipqe_ip->ip_len) - |
ipqe->ipqe_ip->ip_off; |
ntohs(ipqe->ipqe_ip->ip_off); |
if (i > 0) { |
if (i > 0) { |
if (i >= ipqe->ipqe_ip->ip_len) |
if (i >= ntohs(ipqe->ipqe_ip->ip_len)) |
goto dropfrag; |
goto dropfrag; |
m_adj(ipqe->ipqe_m, i); |
m_adj(ipqe->ipqe_m, i); |
ipqe->ipqe_ip->ip_off += i; |
ipqe->ipqe_ip->ip_off = |
ipqe->ipqe_ip->ip_len -= i; |
htons(ntohs(ipqe->ipqe_ip->ip_off) + i); |
|
ipqe->ipqe_ip->ip_len = |
|
htons(ntohs(ipqe->ipqe_ip->ip_len) - i); |
} |
} |
} |
} |
|
|
Line 803 ip_reass(ipqe, fp) |
|
Line 1039 ip_reass(ipqe, fp) |
|
* While we overlap succeeding segments trim them or, |
* While we overlap succeeding segments trim them or, |
* if they are completely covered, dequeue them. |
* if they are completely covered, dequeue them. |
*/ |
*/ |
for (; q != NULL && ipqe->ipqe_ip->ip_off + ipqe->ipqe_ip->ip_len > |
for (; q != NULL && |
q->ipqe_ip->ip_off; q = nq) { |
ntohs(ipqe->ipqe_ip->ip_off) + ntohs(ipqe->ipqe_ip->ip_len) > |
i = (ipqe->ipqe_ip->ip_off + ipqe->ipqe_ip->ip_len) - |
ntohs(q->ipqe_ip->ip_off); q = nq) { |
q->ipqe_ip->ip_off; |
i = (ntohs(ipqe->ipqe_ip->ip_off) + |
if (i < q->ipqe_ip->ip_len) { |
ntohs(ipqe->ipqe_ip->ip_len)) - ntohs(q->ipqe_ip->ip_off); |
q->ipqe_ip->ip_len -= i; |
if (i < ntohs(q->ipqe_ip->ip_len)) { |
q->ipqe_ip->ip_off += i; |
q->ipqe_ip->ip_len = |
|
htons(ntohs(q->ipqe_ip->ip_len) - i); |
|
q->ipqe_ip->ip_off = |
|
htons(ntohs(q->ipqe_ip->ip_off) + i); |
m_adj(q->ipqe_m, i); |
m_adj(q->ipqe_m, i); |
break; |
break; |
} |
} |
nq = q->ipqe_q.le_next; |
nq = TAILQ_NEXT(q, ipqe_q); |
m_freem(q->ipqe_m); |
m_freem(q->ipqe_m); |
LIST_REMOVE(q, ipqe_q); |
TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); |
pool_put(&ipqent_pool, q); |
pool_put(&ipqent_pool, q); |
} |
} |
|
|
|
|
* check for complete reassembly. |
* check for complete reassembly. |
*/ |
*/ |
if (p == NULL) { |
if (p == NULL) { |
LIST_INSERT_HEAD(&fp->ipq_fragq, ipqe, ipqe_q); |
TAILQ_INSERT_HEAD(&fp->ipq_fragq, ipqe, ipqe_q); |
} else { |
} else { |
LIST_INSERT_AFTER(p, ipqe, ipqe_q); |
TAILQ_INSERT_AFTER(&fp->ipq_fragq, p, ipqe, ipqe_q); |
} |
} |
next = 0; |
next = 0; |
for (p = NULL, q = fp->ipq_fragq.lh_first; q != NULL; |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = q->ipqe_q.le_next) { |
p = q, q = TAILQ_NEXT(q, ipqe_q)) { |
if (q->ipqe_ip->ip_off != next) |
if (ntohs(q->ipqe_ip->ip_off) != next) |
return (0); |
return (0); |
next += q->ipqe_ip->ip_len; |
next += ntohs(q->ipqe_ip->ip_len); |
} |
} |
if (p->ipqe_mff) |
if (p->ipqe_mff) |
return (0); |
return (0); |
|
|
* 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 = TAILQ_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 = TAILQ_NEXT(q, ipqe_q); |
pool_put(&ipqent_pool, q); |
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 = TAILQ_NEXT(q, ipqe_q); |
pool_put(&ipqent_pool, q); |
pool_put(&ipqent_pool, q); |
m_cat(m, t); |
m_cat(m, t); |
} |
} |
|
|
* dequeue and discard fragment reassembly header. |
* dequeue and discard fragment reassembly header. |
* Make header visible. |
* Make header visible. |
*/ |
*/ |
ip->ip_len = next; |
ip->ip_len = htons(next); |
ip->ip_src = fp->ipq_src; |
ip->ip_src = fp->ipq_src; |
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 */ |
|
|
|
|
IPQ_LOCK_CHECK(); |
IPQ_LOCK_CHECK(); |
|
|
for (q = fp->ipq_fragq.lh_first; q != NULL; q = p) { |
for (q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; q = p) { |
p = q->ipqe_q.le_next; |
p = TAILQ_NEXT(q, ipqe_q); |
m_freem(q->ipqe_m); |
m_freem(q->ipqe_m); |
LIST_REMOVE(q, ipqe_q); |
TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); |
pool_put(&ipqent_pool, q); |
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--; |
} |
} |
|
|
/* |
/* |
|
|
int s = splsoftnet(); |
int s = splsoftnet(); |
|
|
IPQ_LOCK(); |
IPQ_LOCK(); |
for (fp = ipq.lh_first; fp != NULL; fp = nfp) { |
for (fp = LIST_FIRST(&ipq); fp != NULL; fp = nfp) { |
nfp = fp->ipq_q.le_next; |
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(); |
IPQ_UNLOCK(); |
#ifdef GATEWAY |
#ifdef GATEWAY |
ipflow_slowtimo(); |
ipflow_slowtimo(); |
|
|
if (ipq_lock_try() == 0) |
if (ipq_lock_try() == 0) |
return; |
return; |
|
|
while (ipq.lh_first != NULL) { |
while (LIST_FIRST(&ipq) != NULL) { |
ipstat.ips_fragdropped++; |
ipstat.ips_fragdropped++; |
ip_freef(ipq.lh_first); |
ip_freef(LIST_FIRST(&ipq)); |
} |
} |
|
|
IPQ_UNLOCK(); |
IPQ_UNLOCK(); |
|
|
if (m == 0) |
if (m == 0) |
return ((struct mbuf *)0); |
return ((struct mbuf *)0); |
|
|
|
MCLAIM(m, &inetdomain.dom_mowner); |
#define OPTSIZ (sizeof(ip_srcrt.nop) + sizeof(ip_srcrt.srcopt)) |
#define OPTSIZ (sizeof(ip_srcrt.nop) + sizeof(ip_srcrt.srcopt)) |
|
|
/* length is (nhops+1)*sizeof(addr) + sizeof(nop + srcrt header) */ |
/* length is (nhops+1)*sizeof(addr) + sizeof(nop + srcrt header) */ |
Line 1339 ip_stripoptions(m, mopt) |
|
Line 1592 ip_stripoptions(m, mopt) |
|
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_len -= olen; |
ip->ip_len = htons(ntohs(ip->ip_len) - olen); |
ip->ip_hl = sizeof (struct ip) >> 2; |
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 1378 ip_forward(m, srcrt) |
|
Line 1631 ip_forward(m, srcrt) |
|
struct mbuf *mcopy; |
struct mbuf *mcopy; |
n_long dest; |
n_long dest; |
struct ifnet *destifp; |
struct ifnet *destifp; |
#ifdef IPSEC |
#if defined(IPSEC) || defined(FAST_IPSEC) |
struct ifnet dummyifp; |
struct ifnet dummyifp; |
#endif |
#endif |
|
|
|
/* |
|
* We are now in the output path. |
|
*/ |
|
MCLAIM(m, &ip_tx_mowner); |
|
|
|
/* |
|
* 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) |
Line 1424 ip_forward(m, srcrt) |
|
Line 1687 ip_forward(m, srcrt) |
|
* 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. |
* Pullup to avoid sharing mbuf cluster between m and mcopy. |
*/ |
*/ |
mcopy = m_copym(m, 0, imin((int)ip->ip_len, 68), M_DONTWAIT); |
mcopy = m_copym(m, 0, imin(ntohs(ip->ip_len), 68), M_DONTWAIT); |
if (mcopy) |
if (mcopy) |
mcopy = m_pullup(mcopy, ip->ip_hl << 2); |
mcopy = m_pullup(mcopy, ip->ip_hl << 2); |
|
|
Line 1461 ip_forward(m, srcrt) |
|
Line 1724 ip_forward(m, srcrt) |
|
} |
} |
} |
} |
|
|
#ifdef IPSEC |
|
/* Don't lookup socket in forwading 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)), |
|
(struct ip_moptions *)NULL, (struct socket *)NULL); |
|
|
if (error) |
if (error) |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
else { |
else { |
Line 1506 ip_forward(m, srcrt) |
|
Line 1767 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 !defined(IPSEC) && !defined(FAST_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 |
#else |
Line 1523 ip_forward(m, srcrt) |
|
Line 1784 ip_forward(m, srcrt) |
|
struct route *ro; |
struct route *ro; |
|
|
sp = ipsec4_getpolicybyaddr(mcopy, |
sp = ipsec4_getpolicybyaddr(mcopy, |
IPSEC_DIR_OUTBOUND, |
IPSEC_DIR_OUTBOUND, IP_FORWARDING, |
IP_FORWARDING, |
&ipsecerror); |
&ipsecerror); |
|
|
|
if (sp == NULL) |
if (sp == NULL) |
destifp = ipforward_rt.ro_rt->rt_ifp; |
destifp = ipforward_rt.ro_rt->rt_ifp; |
else { |
else { |
/* count IPsec header size */ |
/* count IPsec header size */ |
ipsechdr = ipsec4_hdrsiz(mcopy, |
ipsechdr = ipsec4_hdrsiz(mcopy, |
IPSEC_DIR_OUTBOUND, |
IPSEC_DIR_OUTBOUND, NULL); |
NULL); |
|
|
|
/* |
/* |
* find the correct route for outer IPv4 |
* find the correct route for outer IPv4 |
Line 1551 ip_forward(m, srcrt) |
|
Line 1810 ip_forward(m, srcrt) |
|
ro = &sp->req->sav->sah->sa_route; |
ro = &sp->req->sav->sah->sa_route; |
if (ro->ro_rt && ro->ro_rt->rt_ifp) { |
if (ro->ro_rt && ro->ro_rt->rt_ifp) { |
dummyifp.if_mtu = |
dummyifp.if_mtu = |
|
ro->ro_rt->rt_rmx.rmx_mtu ? |
|
ro->ro_rt->rt_rmx.rmx_mtu : |
ro->ro_rt->rt_ifp->if_mtu; |
ro->ro_rt->rt_ifp->if_mtu; |
dummyifp.if_mtu -= ipsechdr; |
dummyifp.if_mtu -= ipsechdr; |
destifp = &dummyifp; |
destifp = &dummyifp; |
} |
} |
} |
} |
|
|
|
#ifdef IPSEC |
key_freesp(sp); |
key_freesp(sp); |
|
#else |
|
KEY_FREESP(&sp); |
|
#endif |
} |
} |
} |
} |
#endif /*IPSEC*/ |
#endif /*IPSEC*/ |
Line 1565 ip_forward(m, srcrt) |
|
Line 1830 ip_forward(m, srcrt) |
|
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 if 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); |
} |
} |
Line 1681 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1958 ip_sysctl(name, namelen, oldp, oldlenp, |
|
case IPCTL_MTUDISC: |
case IPCTL_MTUDISC: |
error = sysctl_int(oldp, oldlenp, newp, newlen, |
error = sysctl_int(oldp, oldlenp, newp, newlen, |
&ip_mtudisc); |
&ip_mtudisc); |
if (ip_mtudisc != 0 && ip_mtudisc_timeout_q == NULL) { |
if (error == 0 && ip_mtudisc == 0) |
ip_mtudisc_timeout_q = |
rt_timer_queue_remove_all(ip_mtudisc_timeout_q, TRUE); |
rt_timer_queue_create(ip_mtudisc_timeout); |
|
} else if (ip_mtudisc == 0 && ip_mtudisc_timeout_q != NULL) { |
|
rt_timer_queue_destroy(ip_mtudisc_timeout_q, TRUE); |
|
ip_mtudisc_timeout_q = NULL; |
|
} |
|
return error; |
return error; |
case IPCTL_ANONPORTMIN: |
case IPCTL_ANONPORTMIN: |
old = anonportmin; |
old = anonportmin; |
Line 1716 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1988 ip_sysctl(name, namelen, oldp, oldlenp, |
|
} |
} |
return (error); |
return (error); |
case IPCTL_MTUDISCTIMEOUT: |
case IPCTL_MTUDISCTIMEOUT: |
|
old = ip_mtudisc_timeout; |
error = sysctl_int(oldp, oldlenp, newp, newlen, |
error = sysctl_int(oldp, oldlenp, newp, newlen, |
&ip_mtudisc_timeout); |
&ip_mtudisc_timeout); |
if (ip_mtudisc_timeout_q != NULL) |
if (ip_mtudisc_timeout < 0) { |
rt_timer_queue_change(ip_mtudisc_timeout_q, |
ip_mtudisc_timeout = old; |
|
return (EINVAL); |
|
} |
|
if (error == 0) |
|
rt_timer_queue_change(ip_mtudisc_timeout_q, |
ip_mtudisc_timeout); |
ip_mtudisc_timeout); |
return (error); |
return (error); |
#ifdef GATEWAY |
#ifdef GATEWAY |
Line 1740 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 2017 ip_sysctl(name, namelen, oldp, oldlenp, |
|
&hostzeroisbroadcast)); |
&hostzeroisbroadcast)); |
#if NGIF > 0 |
#if NGIF > 0 |
case IPCTL_GIF_TTL: |
case IPCTL_GIF_TTL: |
return(sysctl_int(oldp, oldlenp, newp, newlen, |
return (sysctl_int(oldp, oldlenp, newp, newlen, |
&ip_gif_ttl)); |
&ip_gif_ttl)); |
#endif |
#endif |
|
|
|
#if NGRE > 0 |
|
case IPCTL_GRE_TTL: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_gre_ttl)); |
|
#endif |
|
|
#ifndef IPNOPRIVPORTS |
#ifndef IPNOPRIVPORTS |
case IPCTL_LOWPORTMIN: |
case IPCTL_LOWPORTMIN: |
old = lowportmin; |
old = lowportmin; |
Line 1769 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 2052 ip_sysctl(name, namelen, oldp, oldlenp, |
|
return (error); |
return (error); |
#endif |
#endif |
|
|
|
case IPCTL_MAXFRAGPACKETS: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_maxfragpackets)); |
|
|
|
case IPCTL_CHECKINTERFACE: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_checkinterface)); |
default: |
default: |
return (EOPNOTSUPP); |
return (EOPNOTSUPP); |
} |
} |