version 1.27, 1996/01/16 04:17:37 |
version 1.245.2.5, 2007/10/09 13:44:50 |
|
|
/* $NetBSD$ */ |
/* $NetBSD$ */ |
|
|
/* |
/* |
* Copyright (c) 1982, 1986, 1988, 1993 |
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. |
* The Regents of the University of California. All rights reserved. |
* 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. |
|
* All rights reserved. |
|
* |
|
* This code is derived from software contributed to The NetBSD Foundation |
|
* by Public Access Networks Corporation ("Panix"). It was developed under |
|
* contract to Panix by Eric Haszlakiewicz and Thor Lancelot Simon. |
* |
* |
* 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 |
|
|
* 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. All advertising materials mentioning features or use of this software |
* must display the following acknowledgement: |
* must display the following acknowledgement: |
* This product includes software developed by the University of |
* This product includes software developed by the NetBSD |
* California, Berkeley and its contributors. |
* Foundation, Inc. and its contributors. |
* 4. Neither the name of the University nor the names of its contributors |
* 4. Neither the name of The NetBSD Foundation 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 NETBSD FOUNDATION, INC. 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 FOUNDATION OR CONTRIBUTORS |
|
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR |
|
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF |
|
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS |
|
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN |
|
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) |
|
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
|
* POSSIBILITY OF SUCH DAMAGE. |
|
*/ |
|
|
|
/* |
|
* Copyright (c) 1982, 1986, 1988, 1993 |
|
* The Regents of the University of California. All rights reserved. |
|
* |
|
* Redistribution and use in source and binary forms, with or without |
|
* modification, are permitted provided that the following conditions |
|
* 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 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_inet.h" |
|
#include "opt_gateway.h" |
|
#include "opt_pfil_hooks.h" |
|
#include "opt_ipsec.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/malloc.h> |
#include <sys/malloc.h> |
|
|
#include <sys/domain.h> |
#include <sys/domain.h> |
#include <sys/protosw.h> |
#include <sys/protosw.h> |
#include <sys/socket.h> |
#include <sys/socket.h> |
|
#include <sys/socketvar.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/pool.h> |
|
#include <sys/sysctl.h> |
|
#include <sys/kauth.h> |
|
|
#include <net/if.h> |
#include <net/if.h> |
|
#include <net/if_dl.h> |
#include <net/route.h> |
#include <net/route.h> |
|
#include <net/pfil.h> |
|
|
#include <netinet/in.h> |
#include <netinet/in.h> |
#include <netinet/in_systm.h> |
#include <netinet/in_systm.h> |
#include <netinet/ip.h> |
#include <netinet/ip.h> |
#include <netinet/in_pcb.h> |
#include <netinet/in_pcb.h> |
|
#include <netinet/in_proto.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 |
|
#ifdef FAST_IPSEC |
|
#include <netipsec/ipsec.h> |
|
#include <netipsec/key.h> |
|
#endif /* FAST_IPSEC*/ |
|
|
#ifndef IPFORWARDING |
#ifndef IPFORWARDING |
#ifdef GATEWAY |
#ifdef GATEWAY |
|
|
#define IPSENDREDIRECTS 1 |
#define IPSENDREDIRECTS 1 |
#endif |
#endif |
#ifndef IPFORWSRCRT |
#ifndef IPFORWSRCRT |
#define IPFORWSRCRT 1 /* allow source-routed packets */ |
#define IPFORWSRCRT 1 /* forward source-routed packets */ |
|
#endif |
|
#ifndef IPALLOWSRCRT |
|
#define IPALLOWSRCRT 1 /* allow source-routed packets */ |
#endif |
#endif |
|
#ifndef IPMTUDISC |
|
#define IPMTUDISC 1 |
|
#endif |
|
#ifndef IPMTUDISCTIMEOUT |
|
#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ |
|
#endif |
|
|
/* |
/* |
* Note: DIRECTED_BROADCAST is handled this way so that previous |
* Note: DIRECTED_BROADCAST is handled this way so that previous |
* configuration using this option will Just Work. |
* configuration using this option will Just Work. |
Line 86 int ipsendredirects = IPSENDREDIRECTS; |
|
Line 194 int ipsendredirects = IPSENDREDIRECTS; |
|
int ip_defttl = IPDEFTTL; |
int ip_defttl = IPDEFTTL; |
int ip_forwsrcrt = IPFORWSRCRT; |
int ip_forwsrcrt = IPFORWSRCRT; |
int ip_directedbcast = IPDIRECTEDBCAST; |
int ip_directedbcast = IPDIRECTEDBCAST; |
|
int ip_allowsrcrt = IPALLOWSRCRT; |
|
int ip_mtudisc = IPMTUDISC; |
|
int ip_mtudisc_timeout = IPMTUDISCTIMEOUT; |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
int ipprintfs = 0; |
int ipprintfs = 0; |
#endif |
#endif |
|
|
extern struct domain inetdomain; |
int ip_do_randomid = 0; |
extern struct protosw inetsw[]; |
|
u_char ip_protox[IPPROTO_MAX]; |
/* |
|
* 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; |
|
|
int ipqmaxlen = IFQ_MAXLEN; |
int ipqmaxlen = IFQ_MAXLEN; |
struct in_ifaddrhead in_ifaddr; |
u_long in_ifaddrhash; /* size of hash table - 1 */ |
|
int in_ifaddrentries; /* total number of addrs */ |
|
struct in_ifaddrhead in_ifaddrhead; |
|
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; |
|
uint16_t ip_id; |
|
|
|
#ifdef PFIL_HOOKS |
|
struct pfil_head inet_pfil_hook; |
|
#endif |
|
|
|
/* |
|
* Cached copy of nmbclusters. If nbclusters is different, |
|
* recalculate IP parameters derived from nmbclusters. |
|
*/ |
|
static int ip_nmbclusters; /* copy of nmbclusters */ |
|
static void ip_nmbclusters_changed(void); /* recalc limits */ |
|
|
|
#define CHECK_NMBCLUSTER_PARAMS() \ |
|
do { \ |
|
if (__predict_false(ip_nmbclusters != nmbclusters)) \ |
|
ip_nmbclusters_changed(); \ |
|
} while (/*CONSTCOND*/0) |
|
|
|
/* IP datagram reassembly queues (hashed) */ |
|
#define IPREASS_NHASH_LOG2 6 |
|
#define IPREASS_NHASH (1 << IPREASS_NHASH_LOG2) |
|
#define IPREASS_HMASK (IPREASS_NHASH - 1) |
|
#define IPREASS_HASH(x,y) \ |
|
(((((x) & 0xF) | ((((x) >> 8) & 0xF) << 4)) ^ (y)) & IPREASS_HMASK) |
|
struct ipqhead ipq[IPREASS_NHASH]; |
|
int ipq_locked; |
|
static int ip_nfragpackets; /* packets in reass queue */ |
|
static int ip_nfrags; /* total fragments in reass queues */ |
|
|
|
int ip_maxfragpackets = 200; /* limit on packets. XXX sysctl */ |
|
int ip_maxfrags; /* limit on fragments. XXX sysctl */ |
|
|
|
|
|
/* |
|
* Additive-Increase/Multiplicative-Decrease (AIMD) strategy for |
|
* IP reassembly queue buffer managment. |
|
* |
|
* We keep a count of total IP fragments (NB: not fragmented packets!) |
|
* awaiting reassembly (ip_nfrags) and a limit (ip_maxfrags) on fragments. |
|
* If ip_nfrags exceeds ip_maxfrags the limit, we drop half the |
|
* total fragments in reassembly queues.This AIMD policy avoids |
|
* repeatedly deleting single packets under heavy fragmentation load |
|
* (e.g., from lossy NFS peers). |
|
*/ |
|
static u_int ip_reass_ttl_decr(u_int ticks); |
|
static void ip_reass_drophalf(void); |
|
|
|
|
|
static inline int ipq_lock_try(void); |
|
static inline void ipq_unlock(void); |
|
|
|
static inline int |
|
ipq_lock_try(void) |
|
{ |
|
int s; |
|
|
|
/* |
|
* Use splvm() -- we're blocking 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(void) |
|
{ |
|
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 (/*CONSTCOND*/ 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 (/*CONSTCOND*/ 0) |
|
#else |
|
#define IPQ_LOCK() (void) ipq_lock_try() |
|
#define IPQ_LOCK_CHECK() /* nothing */ |
|
#endif |
|
|
|
#define IPQ_UNLOCK() ipq_unlock() |
|
|
|
POOL_INIT(inmulti_pool, sizeof(struct in_multi), 0, 0, 0, "inmltpl", NULL, |
|
IPL_SOFTNET); |
|
POOL_INIT(ipqent_pool, sizeof(struct ipqent), 0, 0, 0, "ipqepl", NULL, |
|
IPL_VM); |
|
|
|
#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++ |
|
|
|
EVCNT_ATTACH_STATIC(ip_hwcsum_bad); |
|
EVCNT_ATTACH_STATIC(ip_hwcsum_ok); |
|
EVCNT_ATTACH_STATIC(ip_swcsum); |
|
|
|
#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 112 static struct ip_srcrt { |
|
Line 376 static struct ip_srcrt { |
|
struct in_addr route[MAX_IPOPTLEN/sizeof(struct in_addr)]; |
struct in_addr route[MAX_IPOPTLEN/sizeof(struct in_addr)]; |
} ip_srcrt; |
} ip_srcrt; |
|
|
static void save_rte __P((u_char *, struct in_addr)); |
static void save_rte(u_char *, struct in_addr); |
|
|
|
#ifdef MBUFTRACE |
|
struct mowner ip_rx_mowner = MOWNER_INIT("internet", "rx"); |
|
struct mowner ip_tx_mowner = MOWNER_INIT("internet", "tx"); |
|
#endif |
|
|
|
/* |
|
* Compute IP limits derived from the value of nmbclusters. |
|
*/ |
|
static void |
|
ip_nmbclusters_changed(void) |
|
{ |
|
ip_maxfrags = nmbclusters / 4; |
|
ip_nmbclusters = nmbclusters; |
|
} |
|
|
/* |
/* |
* 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. |
*/ |
*/ |
void |
void |
ip_init() |
ip_init(void) |
{ |
{ |
register struct protosw *pr; |
const struct protosw *pr; |
register int i; |
int i; |
|
|
pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); |
pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW); |
if (pr == 0) |
if (pr == 0) |
|
|
if (pr->pr_domain->dom_family == PF_INET && |
if (pr->pr_domain->dom_family == PF_INET && |
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); |
|
ip_id = time.tv_sec & 0xffff; |
for (i = 0; i < IPREASS_NHASH; i++) |
|
LIST_INIT(&ipq[i]); |
|
|
|
ip_id = time_second & 0xfffff; |
|
|
ipintrq.ifq_maxlen = ipqmaxlen; |
ipintrq.ifq_maxlen = ipqmaxlen; |
TAILQ_INIT(&in_ifaddr); |
ip_nmbclusters_changed(); |
|
|
|
TAILQ_INIT(&in_ifaddrhead); |
|
in_ifaddrhashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IFADDR, |
|
M_WAITOK, &in_ifaddrhash); |
|
in_multihashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IPMADDR, |
|
M_WAITOK, &in_multihash); |
|
ip_mtudisc_timeout_q = rt_timer_queue_create(ip_mtudisc_timeout); |
|
#ifdef GATEWAY |
|
ipflow_init(ip_hashsize); |
|
#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 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 = { |
|
.sin_len = sizeof(ipaddr), |
|
.sin_family = AF_INET, |
|
}; |
struct route ipforward_rt; |
struct route ipforward_rt; |
|
|
/* |
/* |
|
* IP software interrupt routine |
|
*/ |
|
void |
|
ipintr(void) |
|
{ |
|
int s; |
|
struct mbuf *m; |
|
|
|
while (!IF_IS_EMPTY(&ipintrq)) { |
|
s = splnet(); |
|
IF_DEQUEUE(&ipintrq, m); |
|
splx(s); |
|
if (m == 0) |
|
return; |
|
MCLAIM(m, &ip_rx_mowner); |
|
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; |
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; |
struct ipqent *ipqe; |
struct ipqent *ipqe; |
int hlen, mff, s; |
int hlen = 0, mff, len; |
|
int downmatch; |
|
int checkif; |
|
int srcrt = 0; |
|
int s; |
|
u_int hash; |
|
#ifdef FAST_IPSEC |
|
struct m_tag *mtag; |
|
struct tdb_ident *tdbi; |
|
struct secpolicy *sp; |
|
int error; |
|
#endif /* FAST_IPSEC */ |
|
|
next: |
MCLAIM(m, &ip_rx_mowner); |
/* |
|
* 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 |
|
|
/* |
/* |
* 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_ifaddrhead) == 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 |
goto next; |
* 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, void *)) == 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) { |
|
|
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)) { |
|
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. Maybe skip checksum on |
|
* loopback interfaces. |
|
*/ |
|
if (__predict_true(!(m->m_pkthdr.rcvif->if_flags & |
|
IFF_LOOPBACK) || ip_do_loopback_cksum)) { |
|
INET_CSUM_COUNTER_INCR(&ip_swcsum); |
|
if (in_cksum(m, hlen) != 0) |
|
goto badcsum; |
|
} |
|
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) { |
if (ip->ip_len < hlen) { |
ipstat.ips_badlen++; |
ipstat.ips_badlen++; |
|
goto bad; |
goto bad; |
} |
} |
NTOHS(ip->ip_id); |
|
NTOHS(ip->ip_off); |
|
|
|
/* |
/* |
* Check that the amount of data in the buffers |
* Check that the amount of data in the buffers |
|
|
* Trim mbufs if longer than we expect. |
* Trim mbufs if longer than we expect. |
* Drop packet if shorter than we expect. |
* Drop packet if shorter than we expect. |
*/ |
*/ |
if (m->m_pkthdr.len < ip->ip_len) { |
if (m->m_pkthdr.len < len) { |
ipstat.ips_tooshort++; |
ipstat.ips_tooshort++; |
goto bad; |
goto bad; |
} |
} |
if (m->m_pkthdr.len > ip->ip_len) { |
if (m->m_pkthdr.len > len) { |
if (m->m_len == m->m_pkthdr.len) { |
if (m->m_len == m->m_pkthdr.len) { |
m->m_len = ip->ip_len; |
m->m_len = len; |
m->m_pkthdr.len = ip->ip_len; |
m->m_pkthdr.len = len; |
} else |
} else |
m_adj(m, ip->ip_len - m->m_pkthdr.len); |
m_adj(m, len - m->m_pkthdr.len); |
|
} |
|
|
|
#if defined(IPSEC) |
|
/* ipflow (IP fast forwarding) is not compatible with IPsec. */ |
|
m->m_flags &= ~M_CANFASTFWD; |
|
#else |
|
/* |
|
* Assume that we can create a fast-forward IP flow entry |
|
* based on this packet. |
|
*/ |
|
m->m_flags |= M_CANFASTFWD; |
|
#endif |
|
|
|
#ifdef PFIL_HOOKS |
|
/* |
|
* Run through list of hooks for input packets. If there are any |
|
* filters which require that additional packets in the flow are |
|
* not fast-forwarded, they must clear the M_CANFASTFWD flag. |
|
* Note that filters must _never_ set this flag, as another filter |
|
* in the list may have previously cleared it. |
|
*/ |
|
/* |
|
* let ipfilter look at packet on the wire, |
|
* not the decapsulated packet. |
|
*/ |
|
#ifdef IPSEC |
|
if (!ipsec_getnhist(m)) |
|
#elif defined(FAST_IPSEC) |
|
if (!ipsec_indone(m)) |
|
#else |
|
if (1) |
|
#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; |
|
/* |
|
* XXX The setting of "srcrt" here is to prevent ip_forward() |
|
* from generating ICMP redirects for packets that have |
|
* been redirected by a hook back out on to the same LAN that |
|
* they came from and is not an indication that the packet |
|
* is being inffluenced by source routing options. This |
|
* allows things like |
|
* "rdr tlp0 0/0 port 80 -> 1.1.1.200 3128 tcp" |
|
* where tlp0 is both on the 1.1.1.0/24 network and is the |
|
* default route for hosts on 1.1.1.0/24. Of course this |
|
* also requires a "map tlp0 ..." to complete the story. |
|
* One might argue whether or not this kind of network config. |
|
* should be supported in this manner... |
|
*/ |
|
srcrt = (odst.s_addr != ip->ip_dst.s_addr); |
|
} |
|
#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, |
|
|
*/ |
*/ |
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; |
|
|
|
/* |
|
* 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. |
*/ |
* |
for (ia = in_ifaddr.tqh_first; ia; ia = ia->ia_list.tqe_next) { |
* Traditional 4.4BSD did not consult IFF_UP at all. |
if (ip->ip_dst.s_addr == ia->ia_addr.sin_addr.s_addr) |
* The behavior here is to treat addresses on !IFF_UP interface |
goto ours; |
* as not mine. |
if (((ip_directedbcast == 0) || (ip_directedbcast && |
*/ |
ia->ia_ifp == m->m_pkthdr.rcvif)) && |
downmatch = 0; |
(ia->ia_ifp->if_flags & IFF_BROADCAST)) { |
LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { |
if (ip->ip_dst.s_addr == ia->ia_broadaddr.sin_addr.s_addr || |
if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { |
ip->ip_dst.s_addr == ia->ia_netbroadcast.s_addr || |
if (checkif && ia->ia_ifp != m->m_pkthdr.rcvif) |
|
continue; |
|
if ((ia->ia_ifp->if_flags & IFF_UP) != 0) |
|
break; |
|
else |
|
downmatch++; |
|
} |
|
} |
|
if (ia != NULL) |
|
goto ours; |
|
if (m->m_pkthdr.rcvif && m->m_pkthdr.rcvif->if_flags & IFF_BROADCAST) { |
|
IFADDR_FOREACH(ifa, m->m_pkthdr.rcvif) { |
|
if (ifa->ifa_addr->sa_family != AF_INET) |
|
continue; |
|
ia = ifatoia(ifa); |
|
if (in_hosteq(ip->ip_dst, ia->ia_broadaddr.sin_addr) || |
|
in_hosteq(ip->ip_dst, ia->ia_netbroadcast) || |
/* |
/* |
* Look for all-0's host part (old broadcast addr), |
* Look for all-0's host part (old broadcast addr), |
* either for subnet or net. |
* either for subnet or net. |
|
|
ip->ip_dst.s_addr == ia->ia_subnet || |
ip->ip_dst.s_addr == ia->ia_subnet || |
ip->ip_dst.s_addr == ia->ia_net) |
ip->ip_dst.s_addr == ia->ia_net) |
goto ours; |
goto ours; |
|
/* |
|
* An interface with IP address zero accepts |
|
* all packets that arrive on that interface. |
|
*/ |
|
if (in_nullhost(ia->ia_addr.sin_addr)) |
|
goto ours; |
} |
} |
} |
} |
if (IN_MULTICAST(ip->ip_dst.s_addr)) { |
if (IN_MULTICAST(ip->ip_dst.s_addr)) { |
|
|
#ifdef MROUTING |
#ifdef MROUTING |
extern struct socket *ip_mrouter; |
extern struct socket *ip_mrouter; |
|
|
if (m->m_flags & M_EXT) { |
|
if ((m = m_pullup(m, hlen)) == 0) { |
|
ipstat.ips_toosmall++; |
|
goto next; |
|
} |
|
ip = mtod(m, struct ip *); |
|
} |
|
|
|
if (ip_mrouter) { |
if (ip_mrouter) { |
/* |
/* |
* If we are acting as a multicast router, all |
* If we are acting as a multicast router, all |
|
|
* 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 (ip->ip_dst.s_addr == INADDR_BROADCAST || |
if (ip->ip_dst.s_addr == INADDR_BROADCAST || |
ip->ip_dst.s_addr == INADDR_ANY) |
in_nullhost(ip->ip_dst)) |
goto ours; |
goto ours; |
|
|
/* |
/* |
|
|
if (ipforwarding == 0) { |
if (ipforwarding == 0) { |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
m_freem(m); |
m_freem(m); |
} else |
} else { |
ip_forward(m, 0); |
/* |
goto next; |
* 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 |
|
#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; |
|
} |
|
|
|
/* |
|
* Peek at the outbound SP for this packet to determine if |
|
* it's a Fast Forward candidate. |
|
*/ |
|
mtag = m_tag_find(m, PACKET_TAG_IPSEC_PENDING_TDB, NULL); |
|
if (mtag != NULL) |
|
m->m_flags &= ~M_CANFASTFWD; |
|
else { |
|
s = splsoftnet(); |
|
sp = ipsec4_checkpolicy(m, IPSEC_DIR_OUTBOUND, |
|
(IP_FORWARDING | |
|
(ip_directedbcast ? IP_ALLOWBROADCAST : 0)), |
|
&error, NULL); |
|
if (sp != NULL) { |
|
m->m_flags &= ~M_CANFASTFWD; |
|
KEY_FREESP(&sp); |
|
} |
|
splx(s); |
|
} |
|
#endif /* FAST_IPSEC */ |
|
|
|
ip_forward(m, srcrt); |
|
} |
|
return; |
|
|
ours: |
ours: |
/* |
/* |
|
|
* 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) { |
if (ip->ip_off & ~htons(IP_DF|IP_RF)) { |
if (m->m_flags & M_EXT) { /* XXX */ |
|
if ((m = m_pullup(m, sizeof (struct ip))) == 0) { |
|
ipstat.ips_toosmall++; |
|
goto next; |
|
} |
|
ip = mtod(m, struct ip *); |
|
} |
|
/* |
/* |
* 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(); |
|
hash = IPREASS_HASH(ip->ip_src.s_addr, ip->ip_id); |
|
LIST_FOREACH(fp, &ipq[hash], ipq_q) { |
if (ip->ip_id == fp->ipq_id && |
if (ip->ip_id == fp->ipq_id && |
ip->ip_src.s_addr == fp->ipq_src.s_addr && |
in_hosteq(ip->ip_src, fp->ipq_src) && |
ip->ip_dst.s_addr == fp->ipq_dst.s_addr && |
in_hosteq(ip->ip_dst, fp->ipq_dst) && |
ip->ip_p == fp->ipq_p) |
ip->ip_p == fp->ipq_p) |
goto found; |
goto found; |
|
|
|
} |
fp = 0; |
fp = 0; |
found: |
found: |
|
|
|
|
* 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(); |
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++; |
MALLOC(ipqe, struct ipqent *, sizeof (struct ipqent), |
s = splvm(); |
M_IPQ, M_NOWAIT); |
ipqe = pool_get(&ipqent_pool, PR_NOWAIT); |
|
splx(s); |
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_ip = ip; |
ipqe->ipqe_ip = ip; |
ip = ip_reass(ipqe, fp); |
m = ip_reass(ipqe, fp, &ipq[hash]); |
if (ip == 0) |
if (m == 0) { |
goto next; |
IPQ_UNLOCK(); |
|
return; |
|
} |
ipstat.ips_reassembled++; |
ipstat.ips_reassembled++; |
m = dtom(ip); |
ip = mtod(m, struct ip *); |
|
hlen = ip->ip_hl << 2; |
|
ip->ip_len = htons(ntohs(ip->ip_len) + hlen); |
} else |
} else |
if (fp) |
if (fp) |
ip_freef(fp); |
ip_freef(fp); |
} else |
IPQ_UNLOCK(); |
ip->ip_len -= hlen; |
} |
|
|
|
#if defined(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 |
|
#ifdef 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*/ |
|
} |
|
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 (ia && ip) |
|
ia->ia_ifa.ifa_data.ifad_inbytes += ntohs(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); |
} |
} |
|
|
/* |
/* |
|
|
* reassembly of this datagram already exists, then it |
* reassembly of this datagram already exists, then it |
* is given as fp; otherwise have to make a chain. |
* is given as fp; otherwise have to make a chain. |
*/ |
*/ |
struct ip * |
struct mbuf * |
ip_reass(ipqe, fp) |
ip_reass(struct ipqent *ipqe, struct ipq *fp, struct ipqhead *ipqhead) |
register struct ipqent *ipqe; |
|
register struct ipq *fp; |
|
{ |
{ |
register struct mbuf *m = dtom(ipqe->ipqe_ip); |
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, s; |
|
|
|
IPQ_LOCK_CHECK(); |
|
|
/* |
/* |
* Presence of header sizes in mbufs |
* Presence of header sizes in mbufs |
Line 443 ip_reass(ipqe, fp) |
|
Line 1068 ip_reass(ipqe, fp) |
|
m->m_data += hlen; |
m->m_data += hlen; |
m->m_len -= hlen; |
m->m_len -= hlen; |
|
|
|
#ifdef notyet |
|
/* make sure fragment limit is up-to-date */ |
|
CHECK_NMBCLUSTER_PARAMS(); |
|
|
|
/* If we have too many fragments, drop the older half. */ |
|
if (ip_nfrags >= ip_maxfrags) |
|
ip_reass_drophalf(void); |
|
#endif |
|
|
|
/* |
|
* We are about to add a fragment; increment frag count. |
|
*/ |
|
ip_nfrags++; |
|
|
/* |
/* |
* If first fragment to arrive, create a reassembly queue. |
* If first fragment to arrive, create a reassembly queue. |
*/ |
*/ |
if (fp == 0) { |
if (fp == 0) { |
if ((t = m_get(M_DONTWAIT, MT_FTABLE)) == NULL) |
/* |
|
* 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; |
goto dropfrag; |
fp = mtod(t, struct ipq *); |
ip_nfragpackets++; |
LIST_INSERT_HEAD(&ipq, fp, ipq_q); |
MALLOC(fp, struct ipq *, sizeof (struct ipq), |
|
M_FTABLE, M_NOWAIT); |
|
if (fp == NULL) |
|
goto dropfrag; |
|
LIST_INSERT_HEAD(ipqhead, fp, ipq_q); |
|
fp->ipq_nfrags = 1; |
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; |
goto insert; |
goto insert; |
|
} else { |
|
fp->ipq_nfrags++; |
} |
} |
|
|
/* |
/* |
* 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 475 ip_reass(ipqe, fp) |
|
Line 1129 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(dtom(ipqe->ipqe_ip), 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 490 ip_reass(ipqe, fp) |
|
Line 1146 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 = |
m_adj(dtom(q->ipqe_ip), i); |
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); |
break; |
break; |
} |
} |
nq = q->ipqe_q.le_next; |
nq = TAILQ_NEXT(q, ipqe_q); |
m_freem(dtom(q->ipqe_ip)); |
m_freem(q->ipqe_m); |
LIST_REMOVE(q, ipqe_q); |
TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); |
FREE(q, M_IPQ); |
s = splvm(); |
|
pool_put(&ipqent_pool, q); |
|
splx(s); |
|
fp->ipq_nfrags--; |
|
ip_nfrags--; |
} |
} |
|
|
insert: |
insert: |
|
|
* 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; concatenate fragments. |
* Reassembly is complete. Check for a bogus message size and |
|
* concatenate fragments. |
*/ |
*/ |
q = fp->ipq_fragq.lh_first; |
q = TAILQ_FIRST(&fp->ipq_fragq); |
ip = q->ipqe_ip; |
ip = q->ipqe_ip; |
m = dtom(q->ipqe_ip); |
if ((next + (ip->ip_hl << 2)) > IP_MAXPACKET) { |
|
ipstat.ips_toolong++; |
|
ip_freef(fp); |
|
return (0); |
|
} |
|
m = q->ipqe_m; |
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); |
FREE(q, M_IPQ); |
s = splvm(); |
|
pool_put(&ipqent_pool, q); |
|
splx(s); |
for (q = nq; q != NULL; q = nq) { |
for (q = nq; q != NULL; q = nq) { |
t = dtom(q->ipqe_ip); |
t = q->ipqe_m; |
nq = q->ipqe_q.le_next; |
nq = TAILQ_NEXT(q, ipqe_q); |
FREE(q, M_IPQ); |
s = splvm(); |
|
pool_put(&ipqent_pool, q); |
|
splx(s); |
m_cat(m, t); |
m_cat(m, t); |
} |
} |
|
ip_nfrags -= fp->ipq_nfrags; |
|
|
/* |
/* |
* Create header for new ip packet by |
* Create header for new ip packet by |
|
|
* 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); |
(void) m_free(dtom(fp)); |
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; m; m = m->m_next) |
for (t = m; t; t = t->m_next) |
plen += m->m_len; |
plen += t->m_len; |
t->m_pkthdr.len = plen; |
m->m_pkthdr.len = plen; |
|
m->m_pkthdr.csum_flags = 0; |
} |
} |
return (ip); |
return (m); |
|
|
dropfrag: |
dropfrag: |
|
if (fp != 0) |
|
fp->ipq_nfrags--; |
|
ip_nfrags--; |
ipstat.ips_fragdropped++; |
ipstat.ips_fragdropped++; |
m_freem(m); |
m_freem(m); |
FREE(ipqe, M_IPQ); |
s = splvm(); |
|
pool_put(&ipqent_pool, ipqe); |
|
splx(s); |
return (0); |
return (0); |
} |
} |
|
|
|
|
* associated datagrams. |
* associated datagrams. |
*/ |
*/ |
void |
void |
ip_freef(fp) |
ip_freef(struct ipq *fp) |
struct ipq *fp; |
|
{ |
{ |
register struct ipqent *q, *p; |
struct ipqent *q, *p; |
|
u_int nfrags = 0; |
|
int s; |
|
|
|
IPQ_LOCK_CHECK(); |
|
|
|
for (q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; q = p) { |
|
p = TAILQ_NEXT(q, ipqe_q); |
|
m_freem(q->ipqe_m); |
|
nfrags++; |
|
TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); |
|
s = splvm(); |
|
pool_put(&ipqent_pool, q); |
|
splx(s); |
|
} |
|
|
|
if (nfrags != fp->ipq_nfrags) |
|
printf("ip_freef: nfrags %d != %d\n", fp->ipq_nfrags, nfrags); |
|
ip_nfrags -= nfrags; |
|
LIST_REMOVE(fp, ipq_q); |
|
FREE(fp, M_FTABLE); |
|
ip_nfragpackets--; |
|
} |
|
|
|
/* |
|
* IP reassembly TTL machinery for multiplicative drop. |
|
*/ |
|
static u_int fragttl_histo[(IPFRAGTTL+1)]; |
|
|
|
|
|
/* |
|
* Decrement TTL of all reasembly queue entries by `ticks'. |
|
* Count number of distinct fragments (as opposed to partial, fragmented |
|
* datagrams) in the reassembly queue. While we traverse the entire |
|
* reassembly queue, compute and return the median TTL over all fragments. |
|
*/ |
|
static u_int |
|
ip_reass_ttl_decr(u_int ticks) |
|
{ |
|
u_int nfrags, median, dropfraction, keepfraction; |
|
struct ipq *fp, *nfp; |
|
int i; |
|
|
|
nfrags = 0; |
|
memset(fragttl_histo, 0, sizeof fragttl_histo); |
|
|
|
for (i = 0; i < IPREASS_NHASH; i++) { |
|
for (fp = LIST_FIRST(&ipq[i]); fp != NULL; fp = nfp) { |
|
fp->ipq_ttl = ((fp->ipq_ttl <= ticks) ? |
|
0 : fp->ipq_ttl - ticks); |
|
nfp = LIST_NEXT(fp, ipq_q); |
|
if (fp->ipq_ttl == 0) { |
|
ipstat.ips_fragtimeout++; |
|
ip_freef(fp); |
|
} else { |
|
nfrags += fp->ipq_nfrags; |
|
fragttl_histo[fp->ipq_ttl] += fp->ipq_nfrags; |
|
} |
|
} |
|
} |
|
|
for (q = fp->ipq_fragq.lh_first; q != NULL; q = p) { |
KASSERT(ip_nfrags == nfrags); |
p = q->ipqe_q.le_next; |
|
m_freem(dtom(q->ipqe_ip)); |
/* Find median (or other drop fraction) in histogram. */ |
LIST_REMOVE(q, ipqe_q); |
dropfraction = (ip_nfrags / 2); |
FREE(q, M_IPQ); |
keepfraction = ip_nfrags - dropfraction; |
|
for (i = IPFRAGTTL, median = 0; i >= 0; i--) { |
|
median += fragttl_histo[i]; |
|
if (median >= keepfraction) |
|
break; |
} |
} |
LIST_REMOVE(fp, ipq_q); |
|
(void) m_free(dtom(fp)); |
/* Return TTL of median (or other fraction). */ |
|
return (u_int)i; |
|
} |
|
|
|
void |
|
ip_reass_drophalf(void) |
|
{ |
|
|
|
u_int median_ticks; |
|
/* |
|
* Compute median TTL of all fragments, and count frags |
|
* with that TTL or lower (roughly half of all fragments). |
|
*/ |
|
median_ticks = ip_reass_ttl_decr(0); |
|
|
|
/* Drop half. */ |
|
median_ticks = ip_reass_ttl_decr(median_ticks); |
|
|
} |
} |
|
|
/* |
/* |
|
|
* queue, discard it. |
* queue, discard it. |
*/ |
*/ |
void |
void |
ip_slowtimo() |
ip_slowtimo(void) |
{ |
{ |
register struct ipq *fp, *nfp; |
static u_int dropscanidx = 0; |
|
u_int i; |
|
u_int median_ttl; |
int s = splsoftnet(); |
int s = splsoftnet(); |
|
|
for (fp = ipq.lh_first; fp != NULL; fp = nfp) { |
IPQ_LOCK(); |
nfp = fp->ipq_q.le_next; |
|
if (--fp->ipq_ttl == 0) { |
/* Age TTL of all fragments by 1 tick .*/ |
ipstat.ips_fragtimeout++; |
median_ttl = ip_reass_ttl_decr(1); |
ip_freef(fp); |
|
|
/* make sure fragment limit is up-to-date */ |
|
CHECK_NMBCLUSTER_PARAMS(); |
|
|
|
/* If we have too many fragments, drop the older half. */ |
|
if (ip_nfrags > ip_maxfrags) |
|
ip_reass_ttl_decr(median_ttl); |
|
|
|
/* |
|
* If we are over the maximum number of fragmented packets |
|
* (due to the limit being lowered), drain off |
|
* enough to get down to the new limit. Start draining |
|
* from the reassembly hashqueue most recently drained. |
|
*/ |
|
if (ip_maxfragpackets < 0) |
|
; |
|
else { |
|
int wrapped = 0; |
|
|
|
i = dropscanidx; |
|
while (ip_nfragpackets > ip_maxfragpackets && wrapped == 0) { |
|
while (LIST_FIRST(&ipq[i]) != NULL) |
|
ip_freef(LIST_FIRST(&ipq[i])); |
|
if (++i >= IPREASS_NHASH) { |
|
i = 0; |
|
} |
|
/* |
|
* Dont scan forever even if fragment counters are |
|
* wrong: stop after scanning entire reassembly queue. |
|
*/ |
|
if (i == dropscanidx) |
|
wrapped = 1; |
} |
} |
|
dropscanidx = i; |
} |
} |
|
IPQ_UNLOCK(); |
splx(s); |
splx(s); |
} |
} |
|
|
|
|
* Drain off all datagram fragments. |
* Drain off all datagram fragments. |
*/ |
*/ |
void |
void |
ip_drain() |
ip_drain(void) |
{ |
{ |
|
|
while (ipq.lh_first != NULL) { |
/* |
ipstat.ips_fragdropped++; |
* We may be called from a device's interrupt context. If |
ip_freef(ipq.lh_first); |
* the ipq is already busy, just bail out now. |
} |
*/ |
|
if (ipq_lock_try() == 0) |
|
return; |
|
|
|
/* |
|
* Drop half the total fragments now. If more mbufs are needed, |
|
* we will be called again soon. |
|
*/ |
|
ip_reass_drophalf(); |
|
|
|
IPQ_UNLOCK(); |
} |
} |
|
|
/* |
/* |
|
|
* 0 if the packet should be processed further. |
* 0 if the packet should be processed further. |
*/ |
*/ |
int |
int |
ip_dooptions(m) |
ip_dooptions(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 1460 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 1486 ip_dooptions(m) |
|
*/ |
*/ |
case IPOPT_LSRR: |
case IPOPT_LSRR: |
case IPOPT_SSRR: |
case IPOPT_SSRR: |
|
if (ip_allowsrcrt == 0) { |
|
type = ICMP_UNREACH; |
|
code = ICMP_UNREACH_NET_PROHIB; |
|
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 1514 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 1524 ip_dooptions(m) |
|
/* |
/* |
* locate outgoing interface |
* locate outgoing interface |
*/ |
*/ |
bcopy((caddr_t)(cp + off), (caddr_t)&ipaddr.sin_addr, |
bcopy((void *)(cp + off), (void *)&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_ifwithladdr(sintosa(&ipaddr))); |
#define SA struct sockaddr * |
else |
if ((ia = (INA)ifa_ifwithdstaddr((SA)&ipaddr)) == 0) |
|
ia = (INA)ifa_ifwithnet((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 1536 ip_dooptions(m) |
|
goto bad; |
goto bad; |
} |
} |
ip->ip_dst = ipaddr.sin_addr; |
ip->ip_dst = ipaddr.sin_addr; |
bcopy((caddr_t)&ia->ia_addr.sin_addr, |
bcopy((void *)&ia->ia_addr.sin_addr, |
(caddr_t)(cp + off), sizeof(struct in_addr)); |
(void *)(cp + off), sizeof(struct in_addr)); |
cp[IPOPT_OFFSET] += sizeof(struct in_addr); |
cp[IPOPT_OFFSET] += sizeof(struct in_addr); |
/* |
/* |
* Let ip_intr's mcast routing check handle mcast pkts |
* Let ip_intr's mcast routing check handle mcast pkts |
|
Line 1546 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 1558 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((void *)(&ip->ip_dst), (void *)&ipaddr.sin_addr, |
sizeof(ipaddr.sin_addr)); |
sizeof(ipaddr.sin_addr)); |
/* |
/* |
* 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; |
} |
} |
bcopy((caddr_t)&ia->ia_addr.sin_addr, |
bcopy((void *)&ia->ia_addr.sin_addr, |
(caddr_t)(cp + off), sizeof(struct in_addr)); |
(void *)(cp + off), sizeof(struct in_addr)); |
cp[IPOPT_OFFSET] += sizeof(struct in_addr); |
cp[IPOPT_OFFSET] += sizeof(struct in_addr); |
break; |
break; |
|
|
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; |
goto bad; |
|
} |
|
if (ipt->ipt_ptr < 5) { |
|
code = (u_char *)&ipt->ipt_ptr - (u_char *)ip; |
|
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... */ |
|
memmove((char *)cp + ipt->ipt_ptr - 1, cp0, |
sizeof(n_time)); |
sizeof(n_time)); |
ipt->ipt_ptr += sizeof(n_time); |
ipt->ipt_ptr += sizeof(n_time); |
} |
} |
|
Line 1659 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); |
|
|
* return internet address info of interface to be used to get there. |
* return internet address info of interface to be used to get there. |
*/ |
*/ |
struct in_ifaddr * |
struct in_ifaddr * |
ip_rtaddr(dst) |
ip_rtaddr(struct in_addr dst) |
struct in_addr dst; |
|
{ |
{ |
register struct sockaddr_in *sin; |
struct rtentry *rt; |
|
union { |
|
struct sockaddr dst; |
|
struct sockaddr_in dst4; |
|
} u; |
|
|
sin = satosin(&ipforward_rt.ro_dst); |
sockaddr_in_init(&u.dst4, &dst, 0); |
|
|
if (ipforward_rt.ro_rt == 0 || dst.s_addr != sin->sin_addr.s_addr) { |
if ((rt = rtcache_lookup(&ipforward_rt, &u.dst)) == NULL) |
if (ipforward_rt.ro_rt) { |
return NULL; |
RTFREE(ipforward_rt.ro_rt); |
|
ipforward_rt.ro_rt = 0; |
|
} |
|
sin->sin_family = AF_INET; |
|
sin->sin_len = sizeof(*sin); |
|
sin->sin_addr = dst; |
|
|
|
rtalloc(&ipforward_rt); |
return ifatoia(rt->rt_ifa); |
} |
|
if (ipforward_rt.ro_rt == 0) |
|
return ((struct in_ifaddr *)0); |
|
return (ifatoia(ipforward_rt.ro_rt->rt_ifa)); |
|
} |
} |
|
|
/* |
/* |
|
|
* to be picked up later by ip_srcroute if the receiver is interested. |
* to be picked up later by ip_srcroute if the receiver is interested. |
*/ |
*/ |
void |
void |
save_rte(option, dst) |
save_rte(u_char *option, struct in_addr dst) |
u_char *option; |
|
struct in_addr dst; |
|
{ |
{ |
unsigned olen; |
unsigned olen; |
|
|
Line 869 save_rte(option, dst) |
|
Line 1698 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((void *)option, (void *)ip_srcrt.srcopt, olen); |
ip_nhops = (olen - IPOPT_OFFSET - 1) / sizeof(struct in_addr); |
ip_nhops = (olen - IPOPT_OFFSET - 1) / sizeof(struct in_addr); |
ip_srcrt.dst = dst; |
ip_srcrt.dst = dst; |
} |
} |
Line 883 save_rte(option, dst) |
|
Line 1712 save_rte(option, dst) |
|
* The first hop is placed before the options, will be removed later. |
* The first hop is placed before the options, will be removed later. |
*/ |
*/ |
struct mbuf * |
struct mbuf * |
ip_srcroute() |
ip_srcroute(void) |
{ |
{ |
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 NULL; |
m = m_get(M_DONTWAIT, MT_SOOPTS); |
m = m_get(M_DONTWAIT, MT_SOOPTS); |
if (m == 0) |
if (m == 0) |
return ((struct mbuf *)0); |
return NULL; |
|
|
|
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) */ |
|
|
*(mtod(m, struct in_addr *)) = *p--; |
*(mtod(m, struct in_addr *)) = *p--; |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if (ipprintfs) |
if (ipprintfs) |
printf(" hops %lx", ntohl(mtod(m, struct in_addr *)->s_addr)); |
printf(" hops %x", ntohl(mtod(m, struct in_addr *)->s_addr)); |
#endif |
#endif |
|
|
/* |
/* |
|
|
*/ |
*/ |
ip_srcrt.nop = IPOPT_NOP; |
ip_srcrt.nop = IPOPT_NOP; |
ip_srcrt.srcopt[IPOPT_OFFSET] = IPOPT_MINOFF; |
ip_srcrt.srcopt[IPOPT_OFFSET] = IPOPT_MINOFF; |
bcopy((caddr_t)&ip_srcrt.nop, |
memmove(mtod(m, char *) + sizeof(struct in_addr), &ip_srcrt.nop, |
mtod(m, caddr_t) + sizeof(struct in_addr), OPTSIZ); |
OPTSIZ); |
q = (struct in_addr *)(mtod(m, caddr_t) + |
q = (struct in_addr *)(mtod(m, char *) + |
sizeof(struct in_addr) + OPTSIZ); |
sizeof(struct in_addr) + OPTSIZ); |
#undef OPTSIZ |
#undef OPTSIZ |
/* |
/* |
|
|
while (p >= ip_srcrt.route) { |
while (p >= ip_srcrt.route) { |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if (ipprintfs) |
if (ipprintfs) |
printf(" %lx", ntohl(q->s_addr)); |
printf(" %x", ntohl(q->s_addr)); |
#endif |
#endif |
*q++ = *p--; |
*q++ = *p--; |
} |
} |
|
|
*q = ip_srcrt.dst; |
*q = ip_srcrt.dst; |
#ifdef DIAGNOSTIC |
#ifdef DIAGNOSTIC |
if (ipprintfs) |
if (ipprintfs) |
printf(" %lx\n", ntohl(q->s_addr)); |
printf(" %x\n", ntohl(q->s_addr)); |
#endif |
#endif |
return (m); |
return (m); |
} |
} |
|
|
/* |
const int inetctlerrmap[PRC_NCMDS] = { |
* Strip out IP options, at higher |
|
* level protocol in the kernel. |
|
* Second argument is buffer to which options |
|
* will be moved, and return value is their length. |
|
* XXX should be deleted; last arg currently ignored. |
|
*/ |
|
void |
|
ip_stripoptions(m, mopt) |
|
register struct mbuf *m; |
|
struct mbuf *mopt; |
|
{ |
|
register int i; |
|
struct ip *ip = mtod(m, struct ip *); |
|
register caddr_t opts; |
|
int olen; |
|
|
|
olen = (ip->ip_hl<<2) - sizeof (struct ip); |
|
opts = (caddr_t)(ip + 1); |
|
i = m->m_len - (sizeof (struct ip) + olen); |
|
bcopy(opts + olen, opts, (unsigned)i); |
|
m->m_len -= olen; |
|
if (m->m_flags & M_PKTHDR) |
|
m->m_pkthdr.len -= olen; |
|
ip->ip_hl = sizeof(struct ip) >> 2; |
|
} |
|
|
|
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 997 int inetctlerrmap[PRC_NCMDS] = { |
|
Line 1800 int inetctlerrmap[PRC_NCMDS] = { |
|
* via a source route. |
* via a source route. |
*/ |
*/ |
void |
void |
ip_forward(m, srcrt) |
ip_forward(struct mbuf *m, int srcrt) |
struct mbuf *m; |
|
int srcrt; |
|
{ |
{ |
register struct ip *ip = mtod(m, struct ip *); |
struct ip *ip = mtod(m, struct ip *); |
register struct sockaddr_in *sin; |
struct rtentry *rt; |
register struct rtentry *rt; |
int error, type = 0, code = 0, destmtu = 0; |
int error, type = 0, code; |
|
struct mbuf *mcopy; |
struct mbuf *mcopy; |
n_long dest; |
n_long dest; |
struct ifnet *destifp; |
union { |
|
struct sockaddr dst; |
|
struct sockaddr_in dst4; |
|
} u; |
|
|
|
/* |
|
* 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) { |
printf("forward: src %x dst %x ttl %x\n", ip->ip_src, |
printf("forward: src %s ", inet_ntoa(ip->ip_src)); |
ip->ip_dst, ip->ip_ttl); |
printf("dst %s ttl %x\n", inet_ntoa(ip->ip_dst), 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; |
} |
} |
ip->ip_ttl -= IPTTLDEC; |
|
|
|
sin = satosin(&ipforward_rt.ro_dst); |
sockaddr_in_init(&u.dst4, &ip->ip_dst, 0); |
if ((rt = ipforward_rt.ro_rt) == 0 || |
if ((rt = rtcache_lookup(&ipforward_rt, &u.dst)) == NULL) { |
ip->ip_dst.s_addr != sin->sin_addr.s_addr) { |
icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_NET, dest, 0); |
if (ipforward_rt.ro_rt) { |
return; |
RTFREE(ipforward_rt.ro_rt); |
|
ipforward_rt.ro_rt = 0; |
|
} |
|
sin->sin_family = AF_INET; |
|
sin->sin_len = sizeof(*sin); |
|
sin->sin_addr = ip->ip_dst; |
|
|
|
rtalloc(&ipforward_rt); |
|
if (ipforward_rt.ro_rt == 0) { |
|
icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, dest, 0); |
|
return; |
|
} |
|
rt = ipforward_rt.ro_rt; |
|
} |
} |
|
|
/* |
/* |
* Save at most 64 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, 64)); |
mcopy = m_copym(m, 0, imin(ntohs(ip->ip_len), 68), M_DONTWAIT); |
|
if (mcopy) |
|
mcopy = m_pullup(mcopy, ip->ip_hl << 2); |
|
|
|
ip->ip_ttl -= IPTTLDEC; |
|
|
/* |
/* |
* If forwarding packet using same interface that it came in on, |
* If forwarding packet using same interface that it came in on, |
Line 1062 ip_forward(m, srcrt) |
|
Line 1866 ip_forward(m, srcrt) |
|
*/ |
*/ |
if (rt->rt_ifp == m->m_pkthdr.rcvif && |
if (rt->rt_ifp == m->m_pkthdr.rcvif && |
(rt->rt_flags & (RTF_DYNAMIC|RTF_MODIFIED)) == 0 && |
(rt->rt_flags & (RTF_DYNAMIC|RTF_MODIFIED)) == 0 && |
satosin(rt_key(rt))->sin_addr.s_addr != 0 && |
!in_nullhost(satocsin(rt_getkey(rt))->sin_addr) && |
ipsendredirects && !srcrt) { |
ipsendredirects && !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 %lx\n", code, (u_int32_t)dest); |
printf("redirect (%d) to %x\n", code, |
|
(u_int32_t)dest); |
#endif |
#endif |
} |
} |
} |
} |
|
|
error = ip_output(m, (struct mbuf *)0, &ipforward_rt, |
error = ip_output(m, NULL, &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 1090 ip_forward(m, srcrt) |
|
Line 1900 ip_forward(m, srcrt) |
|
if (type) |
if (type) |
ipstat.ips_redirectsent++; |
ipstat.ips_redirectsent++; |
else { |
else { |
if (mcopy) |
if (mcopy) { |
|
#ifdef GATEWAY |
|
if (mcopy->m_flags & M_CANFASTFWD) |
|
ipflow_create(&ipforward_rt, mcopy); |
|
#endif |
m_freem(mcopy); |
m_freem(mcopy); |
|
} |
return; |
return; |
} |
} |
} |
} |
if (mcopy == NULL) |
if (mcopy == NULL) |
return; |
return; |
destifp = NULL; |
|
|
|
switch (error) { |
switch (error) { |
|
|
Line 1117 ip_forward(m, srcrt) |
|
Line 1931 ip_forward(m, srcrt) |
|
case EMSGSIZE: |
case EMSGSIZE: |
type = ICMP_UNREACH; |
type = ICMP_UNREACH; |
code = ICMP_UNREACH_NEEDFRAG; |
code = ICMP_UNREACH_NEEDFRAG; |
if (ipforward_rt.ro_rt) |
#if !defined(IPSEC) && !defined(FAST_IPSEC) |
destifp = ipforward_rt.ro_rt->rt_ifp; |
if (ipforward_rt.ro_rt != NULL) |
|
destmtu = ipforward_rt.ro_rt->rt_ifp->if_mtu; |
|
#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 != NULL) { |
|
struct secpolicy *sp; |
|
int ipsecerror; |
|
size_t ipsechdr; |
|
struct route *ro; |
|
|
|
sp = ipsec4_getpolicybyaddr(mcopy, |
|
IPSEC_DIR_OUTBOUND, IP_FORWARDING, |
|
&ipsecerror); |
|
|
|
if (sp == NULL) |
|
destmtu = ipforward_rt.ro_rt->rt_ifp->if_mtu; |
|
else { |
|
/* count IPsec header size */ |
|
ipsechdr = ipsec4_hdrsiz(mcopy, |
|
IPSEC_DIR_OUTBOUND, NULL); |
|
|
|
/* |
|
* find the correct route for outer IPv4 |
|
* header, compute tunnel MTU. |
|
*/ |
|
|
|
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) { |
|
destmtu = |
|
ro->ro_rt->rt_rmx.rmx_mtu ? |
|
ro->ro_rt->rt_rmx.rmx_mtu : |
|
ro->ro_rt->rt_ifp->if_mtu; |
|
destmtu -= ipsechdr; |
|
} |
|
} |
|
|
|
#ifdef IPSEC |
|
key_freesp(sp); |
|
#else |
|
KEY_FREESP(&sp); |
|
#endif |
|
} |
|
} |
|
#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 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, destmtu); |
} |
} |
|
|
int |
void |
ip_sysctl(name, namelen, oldp, oldlenp, newp, newlen) |
ip_savecontrol(struct inpcb *inp, struct mbuf **mp, struct ip *ip, |
int *name; |
struct mbuf *m) |
u_int namelen; |
{ |
void *oldp; |
|
size_t *oldlenp; |
if (inp->inp_socket->so_options & SO_TIMESTAMP) { |
void *newp; |
struct timeval tv; |
size_t newlen; |
|
{ |
microtime(&tv); |
int temp; |
*mp = sbcreatecontrol((void *) &tv, sizeof(tv), |
|
SCM_TIMESTAMP, SOL_SOCKET); |
/* All sysctl names at this level are terminal. */ |
if (*mp) |
if (namelen != 1) |
mp = &(*mp)->m_next; |
return (ENOTDIR); |
} |
|
if (inp->inp_flags & INP_RECVDSTADDR) { |
switch (name[0]) { |
*mp = sbcreatecontrol((void *) &ip->ip_dst, |
case IPCTL_FORWARDING: |
sizeof(struct in_addr), IP_RECVDSTADDR, IPPROTO_IP); |
return (sysctl_int(oldp, oldlenp, newp, newlen, &ipforwarding)); |
if (*mp) |
case IPCTL_SENDREDIRECTS: |
mp = &(*mp)->m_next; |
return (sysctl_int(oldp, oldlenp, newp, newlen, |
} |
&ipsendredirects)); |
|
case IPCTL_DEFTTL: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_defttl)); |
|
#ifdef notyet |
#ifdef notyet |
case IPCTL_DEFMTU: |
/* |
return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_mtu)); |
* XXX |
|
* Moving these out of udp_input() made them even more broken |
|
* than they already were. |
|
* - fenner@parc.xerox.com |
|
*/ |
|
/* options were tossed already */ |
|
if (inp->inp_flags & INP_RECVOPTS) { |
|
*mp = sbcreatecontrol((void *) opts_deleted_above, |
|
sizeof(struct in_addr), IP_RECVOPTS, IPPROTO_IP); |
|
if (*mp) |
|
mp = &(*mp)->m_next; |
|
} |
|
/* ip_srcroute doesn't do what we want here, need to fix */ |
|
if (inp->inp_flags & INP_RECVRETOPTS) { |
|
*mp = sbcreatecontrol((void *) ip_srcroute(), |
|
sizeof(struct in_addr), IP_RECVRETOPTS, IPPROTO_IP); |
|
if (*mp) |
|
mp = &(*mp)->m_next; |
|
} |
#endif |
#endif |
case IPCTL_FORWSRCRT: |
if (inp->inp_flags & INP_RECVIF) { |
|
struct sockaddr_dl sdl; |
|
|
|
sockaddr_dl_init(&sdl, sizeof(sdl), |
|
(m->m_pkthdr.rcvif != NULL) |
|
? m->m_pkthdr.rcvif->if_index |
|
: 0, |
|
0, NULL, 0, NULL, 0); |
|
*mp = sbcreatecontrol(&sdl, sdl.sdl_len, IP_RECVIF, IPPROTO_IP); |
|
if (*mp) |
|
mp = &(*mp)->m_next; |
|
} |
|
} |
|
|
|
/* |
|
* sysctl helper routine for net.inet.ip.forwsrcrt. |
|
*/ |
|
static int |
|
sysctl_net_inet_ip_forwsrcrt(SYSCTLFN_ARGS) |
|
{ |
|
int error, tmp; |
|
struct sysctlnode node; |
|
|
|
node = *rnode; |
|
tmp = ip_forwsrcrt; |
|
node.sysctl_data = &tmp; |
|
error = sysctl_lookup(SYSCTLFN_CALL(&node)); |
|
if (error || newp == NULL) |
|
return (error); |
|
|
|
if (kauth_authorize_network(l->l_cred, KAUTH_NETWORK_FORWSRCRT, |
|
0, NULL, NULL, NULL)) |
|
return (EPERM); |
|
|
|
ip_forwsrcrt = tmp; |
|
|
|
return (0); |
|
} |
|
|
|
/* |
|
* sysctl helper routine for net.inet.ip.mtudisctimeout. checks the |
|
* range of the new value and tweaks timers if it changes. |
|
*/ |
|
static int |
|
sysctl_net_inet_ip_pmtudto(SYSCTLFN_ARGS) |
|
{ |
|
int error, tmp; |
|
struct sysctlnode node; |
|
|
|
node = *rnode; |
|
tmp = ip_mtudisc_timeout; |
|
node.sysctl_data = &tmp; |
|
error = sysctl_lookup(SYSCTLFN_CALL(&node)); |
|
if (error || newp == NULL) |
|
return (error); |
|
if (tmp < 0) |
|
return (EINVAL); |
|
|
|
ip_mtudisc_timeout = tmp; |
|
rt_timer_queue_change(ip_mtudisc_timeout_q, ip_mtudisc_timeout); |
|
|
|
return (0); |
|
} |
|
|
|
#ifdef GATEWAY |
|
/* |
|
* sysctl helper routine for net.inet.ip.maxflows. |
|
*/ |
|
static int |
|
sysctl_net_inet_ip_maxflows(SYSCTLFN_ARGS) |
|
{ |
|
int s; |
|
|
|
s = sysctl_lookup(SYSCTLFN_CALL(rnode)); |
|
if (s || newp == NULL) |
|
return (s); |
|
|
|
s = splsoftnet(); |
|
ipflow_reap(0); |
|
splx(s); |
|
|
|
return (0); |
|
} |
|
|
|
static int |
|
sysctl_net_inet_ip_hashsize(SYSCTLFN_ARGS) |
|
{ |
|
int error, tmp; |
|
struct sysctlnode node; |
|
|
|
node = *rnode; |
|
tmp = ip_hashsize; |
|
node.sysctl_data = &tmp; |
|
error = sysctl_lookup(SYSCTLFN_CALL(&node)); |
|
if (error || newp == NULL) |
|
return (error); |
|
|
|
if ((tmp & (tmp - 1)) == 0 && tmp != 0) { |
/* |
/* |
* Don't allow this to change in a secure environment. |
* Can only fail due to malloc() |
*/ |
*/ |
if (securelevel > 0) |
if (ipflow_invalidate_all(tmp)) |
return (EPERM); |
return ENOMEM; |
return (sysctl_int(oldp, oldlenp, newp, newlen, |
} else { |
&ip_forwsrcrt)); |
/* |
case IPCTL_DIRECTEDBCAST: |
* EINVAL if not a power of 2 |
return (sysctl_int(oldp, oldlenp, newp, newlen, |
*/ |
&ip_directedbcast)); |
return EINVAL; |
default: |
} |
return (EOPNOTSUPP); |
|
} |
return (0); |
/* NOTREACHED */ |
} |
|
#endif /* GATEWAY */ |
|
|
|
|
|
SYSCTL_SETUP(sysctl_net_inet_ip_setup, "sysctl net.inet.ip subtree setup") |
|
{ |
|
extern int subnetsarelocal, hostzeroisbroadcast; |
|
|
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT, |
|
CTLTYPE_NODE, "net", NULL, |
|
NULL, 0, NULL, 0, |
|
CTL_NET, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT, |
|
CTLTYPE_NODE, "inet", |
|
SYSCTL_DESCR("PF_INET related settings"), |
|
NULL, 0, NULL, 0, |
|
CTL_NET, PF_INET, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT, |
|
CTLTYPE_NODE, "ip", |
|
SYSCTL_DESCR("IPv4 related settings"), |
|
NULL, 0, NULL, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, CTL_EOL); |
|
|
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "forwarding", |
|
SYSCTL_DESCR("Enable forwarding of INET datagrams"), |
|
NULL, 0, &ipforwarding, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_FORWARDING, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "redirect", |
|
SYSCTL_DESCR("Enable sending of ICMP redirect messages"), |
|
NULL, 0, &ipsendredirects, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_SENDREDIRECTS, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "ttl", |
|
SYSCTL_DESCR("Default TTL for an INET datagram"), |
|
NULL, 0, &ip_defttl, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_DEFTTL, CTL_EOL); |
|
#ifdef IPCTL_DEFMTU |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT /* |CTLFLAG_READWRITE? */, |
|
CTLTYPE_INT, "mtu", |
|
SYSCTL_DESCR("Default MTA for an INET route"), |
|
NULL, 0, &ip_mtu, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_DEFMTU, CTL_EOL); |
|
#endif /* IPCTL_DEFMTU */ |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "forwsrcrt", |
|
SYSCTL_DESCR("Enable forwarding of source-routed " |
|
"datagrams"), |
|
sysctl_net_inet_ip_forwsrcrt, 0, &ip_forwsrcrt, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_FORWSRCRT, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "directed-broadcast", |
|
SYSCTL_DESCR("Enable forwarding of broadcast datagrams"), |
|
NULL, 0, &ip_directedbcast, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_DIRECTEDBCAST, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "allowsrcrt", |
|
SYSCTL_DESCR("Accept source-routed datagrams"), |
|
NULL, 0, &ip_allowsrcrt, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_ALLOWSRCRT, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "subnetsarelocal", |
|
SYSCTL_DESCR("Whether logical subnets are considered " |
|
"local"), |
|
NULL, 0, &subnetsarelocal, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_SUBNETSARELOCAL, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "mtudisc", |
|
SYSCTL_DESCR("Use RFC1191 Path MTU Discovery"), |
|
NULL, 0, &ip_mtudisc, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MTUDISC, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "anonportmin", |
|
SYSCTL_DESCR("Lowest ephemeral port number to assign"), |
|
sysctl_net_inet_ip_ports, 0, &anonportmin, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_ANONPORTMIN, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "anonportmax", |
|
SYSCTL_DESCR("Highest ephemeral port number to assign"), |
|
sysctl_net_inet_ip_ports, 0, &anonportmax, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_ANONPORTMAX, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "mtudisctimeout", |
|
SYSCTL_DESCR("Lifetime of a Path MTU Discovered route"), |
|
sysctl_net_inet_ip_pmtudto, 0, &ip_mtudisc_timeout, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MTUDISCTIMEOUT, CTL_EOL); |
|
#ifdef GATEWAY |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "maxflows", |
|
SYSCTL_DESCR("Number of flows for fast forwarding"), |
|
sysctl_net_inet_ip_maxflows, 0, &ip_maxflows, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MAXFLOWS, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "hashsize", |
|
SYSCTL_DESCR("Size of hash table for fast forwarding (IPv4)"), |
|
sysctl_net_inet_ip_hashsize, 0, &ip_hashsize, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
CTL_CREATE, CTL_EOL); |
|
#endif /* GATEWAY */ |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "hostzerobroadcast", |
|
SYSCTL_DESCR("All zeroes address is broadcast address"), |
|
NULL, 0, &hostzeroisbroadcast, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_HOSTZEROBROADCAST, CTL_EOL); |
|
#if NGIF > 0 |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "gifttl", |
|
SYSCTL_DESCR("Default TTL for a gif tunnel datagram"), |
|
NULL, 0, &ip_gif_ttl, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_GIF_TTL, CTL_EOL); |
|
#endif /* NGIF */ |
|
#ifndef IPNOPRIVPORTS |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "lowportmin", |
|
SYSCTL_DESCR("Lowest privileged ephemeral port number " |
|
"to assign"), |
|
sysctl_net_inet_ip_ports, 0, &lowportmin, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_LOWPORTMIN, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "lowportmax", |
|
SYSCTL_DESCR("Highest privileged ephemeral port number " |
|
"to assign"), |
|
sysctl_net_inet_ip_ports, 0, &lowportmax, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_LOWPORTMAX, CTL_EOL); |
|
#endif /* IPNOPRIVPORTS */ |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "maxfragpackets", |
|
SYSCTL_DESCR("Maximum number of fragments to retain for " |
|
"possible reassembly"), |
|
NULL, 0, &ip_maxfragpackets, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MAXFRAGPACKETS, CTL_EOL); |
|
#if NGRE > 0 |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "grettl", |
|
SYSCTL_DESCR("Default TTL for a gre tunnel datagram"), |
|
NULL, 0, &ip_gre_ttl, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_GRE_TTL, CTL_EOL); |
|
#endif /* NGRE */ |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "checkinterface", |
|
SYSCTL_DESCR("Enable receive side of Strong ES model " |
|
"from RFC1122"), |
|
NULL, 0, &ip_checkinterface, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_CHECKINTERFACE, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "random_id", |
|
SYSCTL_DESCR("Assign random ip_id values"), |
|
NULL, 0, &ip_do_randomid, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_RANDOMID, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT|CTLFLAG_READWRITE, |
|
CTLTYPE_INT, "do_loopback_cksum", |
|
SYSCTL_DESCR("Perform IP checksum on loopback"), |
|
NULL, 0, &ip_do_loopback_cksum, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_LOOPBACKCKSUM, CTL_EOL); |
|
sysctl_createv(clog, 0, NULL, NULL, |
|
CTLFLAG_PERMANENT, |
|
CTLTYPE_STRUCT, "stats", |
|
SYSCTL_DESCR("IP statistics"), |
|
NULL, 0, &ipstat, sizeof(ipstat), |
|
CTL_NET, PF_INET, IPPROTO_IP, IPCTL_STATS, |
|
CTL_EOL); |
} |
} |