version 1.130.2.6, 2002/01/08 00:34:06 |
version 1.192, 2003/12/08 02:23:27 |
|
|
/* |
/* |
* 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. |
* |
* |
|
|
#include <sys/cdefs.h> |
#include <sys/cdefs.h> |
__KERNEL_RCSID(0, "$NetBSD$"); |
__KERNEL_RCSID(0, "$NetBSD$"); |
|
|
|
#include "opt_inet.h" |
#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 "opt_inet_csum.h" |
|
|
#include <sys/param.h> |
#include <sys/param.h> |
Line 139 __KERNEL_RCSID(0, "$NetBSD$"); |
|
Line 137 __KERNEL_RCSID(0, "$NetBSD$"); |
|
/* 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> |
Line 148 __KERNEL_RCSID(0, "$NetBSD$"); |
|
Line 148 __KERNEL_RCSID(0, "$NetBSD$"); |
|
#include <netinet6/ipsec.h> |
#include <netinet6/ipsec.h> |
#include <netkey/key.h> |
#include <netkey/key.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 |
Line 166 __KERNEL_RCSID(0, "$NetBSD$"); |
|
Line 170 __KERNEL_RCSID(0, "$NetBSD$"); |
|
#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 194 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 |
|
|
|
int ip_do_randomid = 0; |
|
|
|
/* |
|
* 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; |
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; |
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; |
uint16_t ip_id; |
|
|
#ifdef PFIL_HOOKS |
#ifdef PFIL_HOOKS |
struct pfil_head inet_pfil_hook; |
struct pfil_head inet_pfil_hook; |
#endif |
#endif |
|
|
struct ipqhead ipq; |
/* 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; |
int ipq_locked; |
int ip_nfragpackets = 0; |
int ip_nfragpackets = 0; |
int ip_maxfragpackets = 200; |
int ip_maxfragpackets = 200; |
|
int ip_nfrags = 0; /* total fragments in reass queues */ |
|
|
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; |
|
|
/* |
/* |
* Use splvm() -- we're bloking things that would cause |
* Use splvm() -- we're blocking things that would cause |
* mbuf allocation. |
* mbuf allocation. |
*/ |
*/ |
s = splvm(); |
s = splvm(); |
|
|
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 |
#ifdef INET_CSUM_COUNTERS |
Line 305 static struct ip_srcrt { |
|
Line 340 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) |
|
|
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.tv_sec & 0xfffff; |
ipintrq.ifq_maxlen = ipqmaxlen; |
ipintrq.ifq_maxlen = ipqmaxlen; |
TAILQ_INIT(&in_ifaddr); |
TAILQ_INIT(&in_ifaddrhead); |
in_ifaddrhashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, M_IFADDR, |
in_ifaddrhashtbl = hashinit(IN_IFADDR_HASH_SIZE, HASH_LIST, 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 |
|
|
evcnt_attach_static(&ip_hwcsum_ok); |
evcnt_attach_static(&ip_hwcsum_ok); |
evcnt_attach_static(&ip_swcsum); |
evcnt_attach_static(&ip_swcsum); |
#endif /* INET_CSUM_COUNTERS */ |
#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 }; |
|
|
splx(s); |
splx(s); |
if (m == 0) |
if (m == 0) |
return; |
return; |
|
MCLAIM(m, &ip_rx_mowner); |
ip_input(m); |
ip_input(m); |
} |
} |
} |
} |
Line 394 ip_input(struct mbuf *m) |
|
Line 445 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; |
|
int checkif; |
|
int srcrt = 0; |
|
u_int hash; |
|
#ifdef FAST_IPSEC |
|
struct m_tag *mtag; |
|
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 (TAILQ_FIRST(&in_ifaddr) == 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 |
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 506 ip_input(struct mbuf *m) |
|
Line 572 ip_input(struct mbuf *m) |
|
m_adj(m, len - m->m_pkthdr.len); |
m_adj(m, len - m->m_pkthdr.len); |
} |
} |
|
|
#ifdef IPSEC |
#if defined(IPSEC) || defined(FAST_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 531 ip_input(struct mbuf *m) |
|
Line 597 ip_input(struct mbuf *m) |
|
*/ |
*/ |
#ifdef IPSEC |
#ifdef IPSEC |
if (!ipsec_getnhist(m)) |
if (!ipsec_getnhist(m)) |
|
#elif defined(FAST_IPSEC) |
|
if (!ipsec_indone(m)) |
#else |
#else |
if (1) |
if (1) |
#endif |
#endif |
{ |
{ |
|
struct in_addr odst; |
|
|
|
odst = ip->ip_dst; |
if (pfil_run_hooks(&inet_pfil_hook, &m, m->m_pkthdr.rcvif, |
if (pfil_run_hooks(&inet_pfil_hook, &m, m->m_pkthdr.rcvif, |
PFIL_IN) != 0) |
PFIL_IN) != 0) |
return; |
return; |
if (m == NULL) |
if (m == NULL) |
return; |
return; |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
hlen = ip->ip_hl << 2; |
hlen = ip->ip_hl << 2; |
|
srcrt = (odst.s_addr != ip->ip_dst.s_addr); |
} |
} |
#endif /* PFIL_HOOKS */ |
#endif /* PFIL_HOOKS */ |
|
|
Line 554 ip_input(struct mbuf *m) |
|
Line 626 ip_input(struct mbuf *m) |
|
#endif |
#endif |
|
|
/* |
/* |
* Convert fields to host representation. |
|
*/ |
|
NTOHS(ip->ip_len); |
|
NTOHS(ip->ip_off); |
|
|
|
/* |
|
* Process options and, if not destined for us, |
* Process options and, if not destined for us, |
* ship it on. ip_dooptions returns 1 when an |
* ship it on. ip_dooptions returns 1 when an |
* error was detected (causing an icmp message |
* error was detected (causing an icmp message |
Line 570 ip_input(struct mbuf *m) |
|
Line 636 ip_input(struct mbuf *m) |
|
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. |
Line 579 ip_input(struct mbuf *m) |
|
Line 665 ip_input(struct mbuf *m) |
|
downmatch = 0; |
downmatch = 0; |
LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { |
LIST_FOREACH(ia, &IN_IFADDR_HASH(ip->ip_dst.s_addr), ia_hash) { |
if (in_hosteq(ia->ia_addr.sin_addr, ip->ip_dst)) { |
if (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 |
Line 614 ip_input(struct mbuf *m) |
|
Line 702 ip_input(struct mbuf *m) |
|
#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; |
Line 685 ip_input(struct mbuf *m) |
|
Line 773 ip_input(struct mbuf *m) |
|
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(); |
LIST_FOREACH(fp, &ipq, ipq_q) |
hash = IPREASS_HASH(ip->ip_src.s_addr, ip->ip_id); |
|
/* XXX LIST_FOREACH(fp, &ipq[hash], ipq_q) */ |
|
for (fp = LIST_FIRST(&ipq[hash]); fp != NULL; |
|
fp = LIST_NEXT(fp, 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) && |
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(); |
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) { |
|
|
ipqe->ipqe_mff = mff; |
ipqe->ipqe_mff = mff; |
ipqe->ipqe_m = m; |
ipqe->ipqe_m = m; |
ipqe->ipqe_ip = ip; |
ipqe->ipqe_ip = ip; |
m = ip_reass(ipqe, fp); |
m = ip_reass(ipqe, fp, &ipq[hash]); |
if (m == 0) { |
if (m == 0) { |
IPQ_UNLOCK(); |
IPQ_UNLOCK(); |
return; |
return; |
|
|
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 |
if (ia && ip) |
if (ia && ip) |
ia->ia_ifa.ifa_data.ifad_inbytes += ip->ip_len; |
ia->ia_ifa.ifa_data.ifad_inbytes += ntohs(ip->ip_len); |
#endif |
#endif |
ipstat.ips_delivered++; |
ipstat.ips_delivered++; |
{ |
{ |
|
|
* is given as fp; otherwise have to make a chain. |
* is given as fp; otherwise have to make a chain. |
*/ |
*/ |
struct mbuf * |
struct mbuf * |
ip_reass(ipqe, fp) |
ip_reass(ipqe, fp, ipqhead) |
struct ipqent *ipqe; |
struct ipqent *ipqe; |
struct ipq *fp; |
struct ipq *fp; |
|
struct ipqhead *ipqhead; |
{ |
{ |
struct mbuf *m = ipqe->ipqe_m; |
struct mbuf *m = ipqe->ipqe_m; |
struct ipqent *nq, *p, *q; |
struct ipqent *nq, *p, *q; |
Line 827 ip_reass(ipqe, fp) |
|
Line 1004 ip_reass(ipqe, fp) |
|
m->m_len -= hlen; |
m->m_len -= hlen; |
|
|
/* |
/* |
|
* 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) { |
Line 845 ip_reass(ipqe, fp) |
|
Line 1027 ip_reass(ipqe, fp) |
|
M_FTABLE, M_NOWAIT); |
M_FTABLE, M_NOWAIT); |
if (fp == NULL) |
if (fp == NULL) |
goto dropfrag; |
goto dropfrag; |
LIST_INSERT_HEAD(&ipq, fp, ipq_q); |
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 = LIST_FIRST(&fp->ipq_fragq); q != NULL; |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = LIST_NEXT(q, ipqe_q)) |
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 870 ip_reass(ipqe, fp) |
|
Line 1055 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 885 ip_reass(ipqe, fp) |
|
Line 1072 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 = LIST_NEXT(q, ipqe_q); |
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); |
|
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 = LIST_FIRST(&fp->ipq_fragq); q != NULL; |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = LIST_NEXT(q, ipqe_q)) { |
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 = LIST_FIRST(&fp->ipq_fragq); |
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 = LIST_NEXT(q, ipqe_q); |
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 = LIST_NEXT(q, ipqe_q); |
nq = TAILQ_NEXT(q, ipqe_q); |
pool_put(&ipqent_pool, q); |
pool_put(&ipqent_pool, q); |
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); |
|
|
return (m); |
return (m); |
|
|
dropfrag: |
dropfrag: |
|
if (fp != 0) |
|
fp->ipq_nfrags--; |
|
ip_nfrags--; |
ipstat.ips_fragdropped++; |
ipstat.ips_fragdropped++; |
m_freem(m); |
m_freem(m); |
pool_put(&ipqent_pool, ipqe); |
pool_put(&ipqent_pool, ipqe); |
|
|
struct ipq *fp; |
struct ipq *fp; |
{ |
{ |
struct ipqent *q, *p; |
struct ipqent *q, *p; |
|
u_int nfrags = 0; |
|
|
IPQ_LOCK_CHECK(); |
IPQ_LOCK_CHECK(); |
|
|
for (q = LIST_FIRST(&fp->ipq_fragq); q != NULL; q = p) { |
for (q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; q = p) { |
p = LIST_NEXT(q, ipqe_q); |
p = TAILQ_NEXT(q, ipqe_q); |
m_freem(q->ipqe_m); |
m_freem(q->ipqe_m); |
LIST_REMOVE(q, ipqe_q); |
nfrags++; |
|
TAILQ_REMOVE(&fp->ipq_fragq, q, ipqe_q); |
pool_put(&ipqent_pool, q); |
pool_put(&ipqent_pool, q); |
} |
} |
|
|
|
if (nfrags != fp->ipq_nfrags) |
|
printf("ip_freef: nfrags %d != %d\n", fp->ipq_nfrags, nfrags); |
|
ip_nfrags -= nfrags; |
LIST_REMOVE(fp, ipq_q); |
LIST_REMOVE(fp, ipq_q); |
FREE(fp, M_FTABLE); |
FREE(fp, M_FTABLE); |
ip_nfragpackets--; |
ip_nfragpackets--; |
|
|
void |
void |
ip_slowtimo() |
ip_slowtimo() |
{ |
{ |
|
static u_int dropscanidx = 0; |
|
u_int i; |
struct ipq *fp, *nfp; |
struct ipq *fp, *nfp; |
int s = splsoftnet(); |
int s = splsoftnet(); |
|
|
IPQ_LOCK(); |
IPQ_LOCK(); |
for (fp = LIST_FIRST(&ipq); fp != NULL; fp = nfp) { |
for (i = 0; i < IPREASS_NHASH; i++) { |
nfp = LIST_NEXT(fp, ipq_q); |
for (fp = LIST_FIRST(&ipq[i]); fp != NULL; fp = nfp) { |
if (--fp->ipq_ttl == 0) { |
nfp = LIST_NEXT(fp, ipq_q); |
ipstat.ips_fragtimeout++; |
if (--fp->ipq_ttl == 0) { |
ip_freef(fp); |
ipstat.ips_fragtimeout++; |
|
ip_freef(fp); |
|
} |
} |
} |
} |
} |
/* |
/* |
* If we are over the maximum number of fragments |
* If we are over the maximum number of fragments |
* (due to the limit being lowered), drain off |
* (due to the limit being lowered), drain off |
* enough to get down to the new limit. |
* enough to get down to the new limit. Start draining |
|
* from the reassembly hashqueue most recently drained. |
*/ |
*/ |
if (ip_maxfragpackets < 0) |
if (ip_maxfragpackets < 0) |
; |
; |
else { |
else { |
while (ip_nfragpackets > ip_maxfragpackets && LIST_FIRST(&ipq)) |
int wrapped = 0; |
ip_freef(LIST_FIRST(&ipq)); |
|
|
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(); |
IPQ_UNLOCK(); |
#ifdef GATEWAY |
#ifdef GATEWAY |
|
|
void |
void |
ip_drain() |
ip_drain() |
{ |
{ |
|
int i; |
|
|
/* |
/* |
* We may be called from a device's interrupt context. If |
* We may be called from a device's interrupt context. If |
|
|
if (ipq_lock_try() == 0) |
if (ipq_lock_try() == 0) |
return; |
return; |
|
|
while (LIST_FIRST(&ipq) != NULL) { |
for (i = 0; i < IPREASS_NHASH; i++) { |
ipstat.ips_fragdropped++; |
struct ipqhead *ipqh = &ipq[i]; |
ip_freef(LIST_FIRST(&ipq)); |
struct ipq *fp, *nfp; |
|
for (fp = LIST_FIRST(ipqh); fp != NULL; fp = nfp) { |
|
nfp = LIST_NEXT(fp, ipq_q); |
|
ip_freef(fp); |
|
ipstat.ips_fragdropped++; |
|
} |
} |
} |
|
|
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 1434 ip_stripoptions(m, mopt) |
|
Line 1663 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; |
} |
} |
|
|
Line 1473 ip_forward(m, srcrt) |
|
Line 1702 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. |
* Clear any in-bound checksum flags for this packet. |
*/ |
*/ |
m->m_pkthdr.csum_flags = 0; |
m->m_pkthdr.csum_flags = 0; |
Line 1524 ip_forward(m, srcrt) |
|
Line 1758 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 1561 ip_forward(m, srcrt) |
|
Line 1795 ip_forward(m, srcrt) |
|
} |
} |
} |
} |
|
|
#ifdef IPSEC |
|
/* Don't lookup socket in forwarding case */ |
|
(void)ipsec_setsocket(m, NULL); |
|
#endif |
|
error = ip_output(m, (struct mbuf *)0, &ipforward_rt, |
error = ip_output(m, (struct mbuf *)0, &ipforward_rt, |
(IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), 0); |
(IP_FORWARDING | (ip_directedbcast ? IP_ALLOWBROADCAST : 0)), |
|
(struct ip_moptions *)NULL, (struct socket *)NULL); |
|
|
if (error) |
if (error) |
ipstat.ips_cantforward++; |
ipstat.ips_cantforward++; |
else { |
else { |
Line 1606 ip_forward(m, srcrt) |
|
Line 1838 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 1623 ip_forward(m, srcrt) |
|
Line 1855 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 1651 ip_forward(m, srcrt) |
|
Line 1881 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 1665 ip_forward(m, srcrt) |
|
Line 1901 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 1732 ip_savecontrol(inp, mp, ip, m) |
|
Line 1980 ip_savecontrol(inp, mp, ip, m) |
|
} |
} |
} |
} |
|
|
int |
/* |
ip_sysctl(name, namelen, oldp, oldlenp, newp, newlen) |
* sysctl helper routine for net.inet.ip.mtudisctimeout. checks the |
int *name; |
* range of the new value and tweaks timers if it changes. |
u_int namelen; |
*/ |
void *oldp; |
static int |
size_t *oldlenp; |
sysctl_net_inet_ip_pmtudto(SYSCTLFN_ARGS) |
void *newp; |
|
size_t newlen; |
|
{ |
{ |
extern int subnetsarelocal, hostzeroisbroadcast; |
int error, tmp; |
|
struct sysctlnode node; |
int error, old; |
|
|
|
/* All sysctl names at this level are terminal. */ |
node = *rnode; |
if (namelen != 1) |
tmp = ip_mtudisc_timeout; |
return (ENOTDIR); |
node.sysctl_data = &tmp; |
|
error = sysctl_lookup(SYSCTLFN_CALL(&node)); |
switch (name[0]) { |
if (error || newp == NULL) |
case IPCTL_FORWARDING: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, &ipforwarding)); |
|
case IPCTL_SENDREDIRECTS: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ipsendredirects)); |
|
case IPCTL_DEFTTL: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_defttl)); |
|
#ifdef notyet |
|
case IPCTL_DEFMTU: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, &ip_mtu)); |
|
#endif |
|
case IPCTL_FORWSRCRT: |
|
/* Don't allow this to change in a secure environment. */ |
|
if (securelevel > 0) |
|
return (sysctl_rdint(oldp, oldlenp, newp, |
|
ip_forwsrcrt)); |
|
else |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_forwsrcrt)); |
|
case IPCTL_DIRECTEDBCAST: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_directedbcast)); |
|
case IPCTL_ALLOWSRCRT: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_allowsrcrt)); |
|
case IPCTL_SUBNETSARELOCAL: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&subnetsarelocal)); |
|
case IPCTL_MTUDISC: |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_mtudisc); |
|
if (ip_mtudisc != 0 && ip_mtudisc_timeout_q == NULL) { |
|
ip_mtudisc_timeout_q = |
|
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; |
|
case IPCTL_ANONPORTMIN: |
|
old = anonportmin; |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmin); |
|
if (anonportmin >= anonportmax || anonportmin < 0 |
|
|| anonportmin > 65535 |
|
#ifndef IPNOPRIVPORTS |
|
|| anonportmin < IPPORT_RESERVED |
|
#endif |
|
) { |
|
anonportmin = old; |
|
return (EINVAL); |
|
} |
|
return (error); |
|
case IPCTL_ANONPORTMAX: |
|
old = anonportmax; |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, &anonportmax); |
|
if (anonportmin >= anonportmax || anonportmax < 0 |
|
|| anonportmax > 65535 |
|
#ifndef IPNOPRIVPORTS |
|
|| anonportmax < IPPORT_RESERVED |
|
#endif |
|
) { |
|
anonportmax = old; |
|
return (EINVAL); |
|
} |
|
return (error); |
|
case IPCTL_MTUDISCTIMEOUT: |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_mtudisc_timeout); |
|
if (ip_mtudisc_timeout_q != NULL) |
|
rt_timer_queue_change(ip_mtudisc_timeout_q, |
|
ip_mtudisc_timeout); |
|
return (error); |
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 |
#ifdef GATEWAY |
case IPCTL_MAXFLOWS: |
/* |
{ |
* sysctl helper routine for net.inet.ip.maxflows. apparently if |
int s; |
* maxflows is even looked up, we "reap flows". |
|
*/ |
|
static int |
|
sysctl_net_inet_ip_maxflows(SYSCTLFN_ARGS) |
|
{ |
|
int s; |
|
|
error = sysctl_int(oldp, oldlenp, newp, newlen, |
s = sysctl_lookup(SYSCTLFN_CALL(rnode)); |
&ip_maxflows); |
if (s) |
s = splsoftnet(); |
return (s); |
ipflow_reap(0); |
|
splx(s); |
s = splsoftnet(); |
return (error); |
ipflow_reap(0); |
} |
splx(s); |
#endif |
|
case IPCTL_HOSTZEROBROADCAST: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&hostzeroisbroadcast)); |
|
#if NGIF > 0 |
|
case IPCTL_GIF_TTL: |
|
return(sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_gif_ttl)); |
|
#endif |
|
|
|
#ifndef IPNOPRIVPORTS |
return (0); |
case IPCTL_LOWPORTMIN: |
} |
old = lowportmin; |
#endif /* GATEWAY */ |
error = sysctl_int(oldp, oldlenp, newp, newlen, &lowportmin); |
|
if (lowportmin >= lowportmax |
|
|| lowportmin > IPPORT_RESERVEDMAX |
|
|| lowportmin < IPPORT_RESERVEDMIN |
|
) { |
|
lowportmin = old; |
|
return (EINVAL); |
|
} |
|
return (error); |
|
case IPCTL_LOWPORTMAX: |
|
old = lowportmax; |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, &lowportmax); |
|
if (lowportmin >= lowportmax |
|
|| lowportmax > IPPORT_RESERVEDMAX |
|
|| lowportmax < IPPORT_RESERVEDMIN |
|
) { |
|
lowportmax = old; |
|
return (EINVAL); |
|
} |
|
return (error); |
|
#endif |
|
|
|
case IPCTL_MAXFRAGPACKETS: |
|
return (sysctl_int(oldp, oldlenp, newp, newlen, |
|
&ip_maxfragpackets)); |
|
|
|
default: |
SYSCTL_SETUP(sysctl_net_inet_ip_setup, "sysctl net.inet.ip subtree setup") |
return (EOPNOTSUPP); |
{ |
} |
extern int subnetsarelocal, hostzeroisbroadcast; |
/* NOTREACHED */ |
|
|
sysctl_createv(SYSCTL_PERMANENT, |
|
CTLTYPE_NODE, "net", NULL, |
|
NULL, 0, NULL, 0, |
|
CTL_NET, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT, |
|
CTLTYPE_NODE, "inet", NULL, |
|
NULL, 0, NULL, 0, |
|
CTL_NET, PF_INET, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT, |
|
CTLTYPE_NODE, "ip", NULL, |
|
NULL, 0, NULL, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, CTL_EOL); |
|
|
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "forwarding", NULL, |
|
NULL, 0, &ipforwarding, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_FORWARDING, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "redirect", NULL, |
|
NULL, 0, &ipsendredirects, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_SENDREDIRECTS, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "ttl", NULL, |
|
NULL, 0, &ip_defttl, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_DEFTTL, CTL_EOL); |
|
#ifdef IPCTL_DEFMTU |
|
sysctl_createv(SYSCTL_PERMANENT /* |SYSCTL_READWRITE? */, |
|
CTLTYPE_INT, "mtu", NULL, |
|
NULL, 0, &ip_mtu, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_DEFMTU, CTL_EOL); |
|
#endif /* IPCTL_DEFMTU */ |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READONLY1, |
|
CTLTYPE_INT, "forwsrcrt", NULL, |
|
NULL, 0, &ip_forwsrcrt, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_FORWSRCRT, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "directed-broadcast", NULL, |
|
NULL, 0, &ip_directedbcast, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_DIRECTEDBCAST, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "allowsrcrt", NULL, |
|
NULL, 0, &ip_allowsrcrt, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_ALLOWSRCRT, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "subnetsarelocal", NULL, |
|
NULL, 0, &subnetsarelocal, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_SUBNETSARELOCAL, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "mtudisc", NULL, |
|
NULL, 0, &ip_mtudisc, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MTUDISC, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "anonportmin", NULL, |
|
sysctl_net_inet_ip_ports, 0, &anonportmin, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_ANONPORTMIN, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "anonportmax", NULL, |
|
sysctl_net_inet_ip_ports, 0, &anonportmax, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_ANONPORTMAX, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "mtudisctimeout", NULL, |
|
sysctl_net_inet_ip_pmtudto, 0, &ip_mtudisc_timeout, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MTUDISCTIMEOUT, CTL_EOL); |
|
#ifdef GATEWAY |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "maxflows", NULL, |
|
sysctl_net_inet_ip_maxflows, 0, &ip_maxflows, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MAXFLOWS, CTL_EOL); |
|
#endif /* GATEWAY */ |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "hostzerobroadcast", NULL, |
|
NULL, 0, &hostzeroisbroadcast, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_HOSTZEROBROADCAST, CTL_EOL); |
|
#if NGIF > 0 |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "gifttl", NULL, |
|
NULL, 0, &ip_gif_ttl, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_GIF_TTL, CTL_EOL); |
|
#endif /* NGIF */ |
|
#ifndef IPNOPRIVPORTS |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "lowportmin", NULL, |
|
sysctl_net_inet_ip_ports, 0, &lowportmin, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_LOWPORTMIN, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "lowportmax", NULL, |
|
sysctl_net_inet_ip_ports, 0, &lowportmax, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_LOWPORTMAX, CTL_EOL); |
|
#endif /* IPNOPRIVPORTS */ |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "maxfragpackets", NULL, |
|
NULL, 0, &ip_maxfragpackets, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_MAXFRAGPACKETS, CTL_EOL); |
|
#if NGRE > 0 |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "grettl", NULL, |
|
NULL, 0, &ip_gre_ttl, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_GRE_TTL, CTL_EOL); |
|
#endif /* NGRE */ |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "checkinterface", NULL, |
|
NULL, 0, &ip_checkinterface, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_CHECKINTERFACE, CTL_EOL); |
|
sysctl_createv(SYSCTL_PERMANENT|SYSCTL_READWRITE, |
|
CTLTYPE_INT, "random_id", NULL, |
|
NULL, 0, &ip_do_randomid, 0, |
|
CTL_NET, PF_INET, IPPROTO_IP, |
|
IPCTL_RANDOMID, CTL_EOL); |
} |
} |