version 1.135.2.5, 2002/09/06 08:49:14 |
version 1.150.4.2, 2002/11/12 14:44:11 |
|
|
/* |
/* |
* 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 |
Line 168 __KERNEL_RCSID(0, "$NetBSD$"); |
|
Line 168 __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 1 |
#define IPMTUDISC 0 |
#endif |
#endif |
#ifndef IPMTUDISCTIMEOUT |
#ifndef IPMTUDISCTIMEOUT |
#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ |
#define IPMTUDISCTIMEOUT (10 * 60) /* as per RFC 1191 */ |
|
|
TAILQ_INIT(&in_ifaddr); |
TAILQ_INIT(&in_ifaddr); |
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) |
ip_mtudisc_timeout_q = 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 |
Line 420 ip_input(struct mbuf *m) |
|
Line 418 ip_input(struct mbuf *m) |
|
if (TAILQ_FIRST(&in_ifaddr) == 0) |
if (TAILQ_FIRST(&in_ifaddr) == 0) |
goto bad; |
goto bad; |
ipstat.ips_total++; |
ipstat.ips_total++; |
/* |
if (m->m_len < sizeof (struct ip) && |
* If the IP header is not aligned, slurp it up into a new |
(m = m_pullup(m, sizeof (struct ip))) == 0) { |
* mbuf with space for link headers, in the event we forward |
ipstat.ips_toosmall++; |
* it. Otherwise, if it is aligned, make sure the entire |
return; |
* 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 572 ip_input(struct mbuf *m) |
|
Line 556 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 |
|
|
* 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 & ~htons(IP_DF|IP_RF)) { |
if (ip->ip_off & ~(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. |
|
|
* 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 = htons(ntohs(ip->ip_len) - hlen); |
ip->ip_len -= hlen; |
mff = (ip->ip_off & htons(IP_MF)) != 0; |
mff = (ip->ip_off & 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 (ntohs(ip->ip_len) == 0 || |
if (ip->ip_len == 0 || (ip->ip_len & 0x7) != 0) { |
(ntohs(ip->ip_len) & 0x7) != 0) { |
|
ipstat.ips_badfrags++; |
ipstat.ips_badfrags++; |
IPQ_UNLOCK(); |
IPQ_UNLOCK(); |
goto bad; |
goto bad; |
} |
} |
} |
} |
ip->ip_off = htons((ntohs(ip->ip_off) & IP_OFFMASK) << 3); |
ip->ip_off <<= 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 != htons(0)) { |
if (mff || ip->ip_off) { |
ipstat.ips_fragments++; |
ipstat.ips_fragments++; |
ipqe = pool_get(&ipqent_pool, PR_NOWAIT); |
ipqe = pool_get(&ipqent_pool, PR_NOWAIT); |
if (ipqe == NULL) { |
if (ipqe == NULL) { |
|
|
ipstat.ips_reassembled++; |
ipstat.ips_reassembled++; |
ip = mtod(m, struct ip *); |
ip = mtod(m, struct ip *); |
hlen = ip->ip_hl << 2; |
hlen = ip->ip_hl << 2; |
ip->ip_len = htons(ntohs(ip->ip_len) + hlen); |
ip->ip_len += hlen; |
} else |
} else |
if (fp) |
if (fp) |
ip_freef(fp); |
ip_freef(fp); |
|
|
*/ |
*/ |
#if IFA_STATS |
#if IFA_STATS |
if (ia && ip) |
if (ia && ip) |
ia->ia_ifa.ifa_data.ifad_inbytes += ntohs(ip->ip_len); |
ia->ia_ifa.ifa_data.ifad_inbytes += ip->ip_len; |
#endif |
#endif |
ipstat.ips_delivered++; |
ipstat.ips_delivered++; |
{ |
{ |
Line 889 ip_reass(ipqe, fp) |
|
Line 870 ip_reass(ipqe, fp) |
|
*/ |
*/ |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = TAILQ_NEXT(q, ipqe_q)) |
p = q, q = TAILQ_NEXT(q, ipqe_q)) |
if (ntohs(q->ipqe_ip->ip_off) > ntohs(ipqe->ipqe_ip->ip_off)) |
if (q->ipqe_ip->ip_off > ipqe->ipqe_ip->ip_off) |
break; |
break; |
|
|
/* |
/* |
Line 898 ip_reass(ipqe, fp) |
|
Line 879 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 = ntohs(p->ipqe_ip->ip_off) + ntohs(p->ipqe_ip->ip_len) - |
i = p->ipqe_ip->ip_off + p->ipqe_ip->ip_len - |
ntohs(ipqe->ipqe_ip->ip_off); |
ipqe->ipqe_ip->ip_off; |
if (i > 0) { |
if (i > 0) { |
if (i >= ntohs(ipqe->ipqe_ip->ip_len)) |
if (i >= 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 = |
ipqe->ipqe_ip->ip_off += i; |
htons(ntohs(ipqe->ipqe_ip->ip_off) + i); |
ipqe->ipqe_ip->ip_len -= i; |
ipqe->ipqe_ip->ip_len = |
|
htons(ntohs(ipqe->ipqe_ip->ip_len) - i); |
|
} |
} |
} |
} |
|
|
Line 915 ip_reass(ipqe, fp) |
|
Line 894 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 && |
for (; q != NULL && ipqe->ipqe_ip->ip_off + ipqe->ipqe_ip->ip_len > |
ntohs(ipqe->ipqe_ip->ip_off) + ntohs(ipqe->ipqe_ip->ip_len) > |
q->ipqe_ip->ip_off; q = nq) { |
ntohs(q->ipqe_ip->ip_off); q = nq) { |
i = (ipqe->ipqe_ip->ip_off + ipqe->ipqe_ip->ip_len) - |
i = (ntohs(ipqe->ipqe_ip->ip_off) + |
q->ipqe_ip->ip_off; |
ntohs(ipqe->ipqe_ip->ip_len)) - ntohs(q->ipqe_ip->ip_off); |
if (i < q->ipqe_ip->ip_len) { |
if (i < ntohs(q->ipqe_ip->ip_len)) { |
q->ipqe_ip->ip_len -= i; |
q->ipqe_ip->ip_len = |
q->ipqe_ip->ip_off += 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); |
m_adj(q->ipqe_m, i); |
break; |
break; |
} |
} |
|
|
next = 0; |
next = 0; |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
for (p = NULL, q = TAILQ_FIRST(&fp->ipq_fragq); q != NULL; |
p = q, q = TAILQ_NEXT(q, ipqe_q)) { |
p = q, q = TAILQ_NEXT(q, ipqe_q)) { |
if (ntohs(q->ipqe_ip->ip_off) != next) |
if (q->ipqe_ip->ip_off != next) |
return (0); |
return (0); |
next += ntohs(q->ipqe_ip->ip_len); |
next += q->ipqe_ip->ip_len; |
} |
} |
if (p->ipqe_mff) |
if (p->ipqe_mff) |
return (0); |
return (0); |
|
|
* dequeue and discard fragment reassembly header. |
* dequeue and discard fragment reassembly header. |
* Make header visible. |
* Make header visible. |
*/ |
*/ |
ip->ip_len = htons(next); |
ip->ip_len = 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); |
Line 1467 ip_stripoptions(m, mopt) |
|
Line 1443 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 = htons(ntohs(ip->ip_len) - olen); |
ip->ip_len -= olen; |
ip->ip_hl = sizeof (struct ip) >> 2; |
ip->ip_hl = sizeof (struct ip) >> 2; |
} |
} |
|
|
Line 1557 ip_forward(m, srcrt) |
|
Line 1533 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(ntohs(ip->ip_len), 68), M_DONTWAIT); |
mcopy = m_copym(m, 0, imin((int)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 1829 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1805 ip_sysctl(name, namelen, oldp, oldlenp, |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, |
error = sysctl_int(oldp, oldlenp, newp, newlen, |
&ip_mtudisc); |
&ip_mtudisc); |
if (ip_mtudisc != 0 && ip_mtudisc_timeout_q == NULL) { |
if (ip_mtudisc != 0 && ip_mtudisc_timeout_q == NULL) { |
ip_mtudisc_timeout_q = |
ip_mtudisc_timeout_q = |
rt_timer_queue_create(ip_mtudisc_timeout); |
rt_timer_queue_create(ip_mtudisc_timeout); |
} else if (ip_mtudisc == 0 && ip_mtudisc_timeout_q != NULL) { |
} else if (ip_mtudisc == 0 && ip_mtudisc_timeout_q != NULL) { |
rt_timer_queue_destroy(ip_mtudisc_timeout_q, TRUE); |
rt_timer_queue_destroy(ip_mtudisc_timeout_q, TRUE); |
Line 1866 ip_sysctl(name, namelen, oldp, oldlenp, |
|
Line 1842 ip_sysctl(name, namelen, oldp, oldlenp, |
|
error = sysctl_int(oldp, oldlenp, newp, newlen, |
error = sysctl_int(oldp, oldlenp, newp, newlen, |
&ip_mtudisc_timeout); |
&ip_mtudisc_timeout); |
if (ip_mtudisc_timeout_q != NULL) |
if (ip_mtudisc_timeout_q != NULL) |
rt_timer_queue_change(ip_mtudisc_timeout_q, |
rt_timer_queue_change(ip_mtudisc_timeout_q, |
ip_mtudisc_timeout); |
ip_mtudisc_timeout); |
return (error); |
return (error); |
#ifdef GATEWAY |
#ifdef GATEWAY |