/*
- * Copyright (c) 2000 Apple Computer, Inc. All rights reserved.
- *
- * @APPLE_LICENSE_HEADER_START@
- *
- * The contents of this file constitute Original Code as defined in and
- * are subject to the Apple Public Source License Version 1.1 (the
- * "License"). You may not use this file except in compliance with the
- * License. Please obtain a copy of the License at
- * http://www.apple.com/publicsource and read it before using this file.
+ * Copyright (c) 2006 Apple Computer, Inc. All Rights Reserved.
*
- * This Original Code and all software distributed under the License are
- * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
- * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
- * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT. Please see the
- * License for the specific language governing rights and limitations
- * under the License.
+ * @APPLE_LICENSE_OSREFERENCE_HEADER_START@
*
- * @APPLE_LICENSE_HEADER_END@
+ * This file contains Original Code and/or Modifications of Original Code
+ * as defined in and that are subject to the Apple Public Source License
+ * Version 2.0 (the 'License'). You may not use this file except in
+ * compliance with the License. The rights granted to you under the
+ * License may not be used to create, or enable the creation or
+ * redistribution of, unlawful or unlicensed copies of an Apple operating
+ * system, or to circumvent, violate, or enable the circumvention or
+ * violation of, any terms of an Apple operating system software license
+ * agreement.
+ *
+ * Please obtain a copy of the License at
+ * http://www.opensource.apple.com/apsl/ and read it before using this
+ * file.
+ *
+ * The Original Code and all software distributed under the License are
+ * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
+ * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
+ * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
+ * Please see the License for the specific language governing rights and
+ * limitations under the License.
+ *
+ * @APPLE_LICENSE_OSREFERENCE_HEADER_END@
*/
/*
* Copyright (c) 1982, 1986, 1988, 1993
* SUCH DAMAGE.
*
* @(#)ip_input.c 8.2 (Berkeley) 1/4/94
- * $ANA: ip_input.c,v 1.5 1996/09/18 14:34:59 wollman Exp $
+ * $FreeBSD: src/sys/netinet/ip_input.c,v 1.130.2.25 2001/08/29 21:41:37 jesper Exp $
*/
#define _IP_VHL
-#include <stddef.h>
-
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/sysctl.h>
#include <kern/queue.h>
+#include <kern/locks.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_dl.h>
#include <net/route.h>
-#include <net/netisr.h>
+#include <net/kpi_protocol.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/ip.h>
-#ifdef INET6
-#include <netinet/ip6.h>
-#include <netinet6/ip6_var.h>
-#endif
#include <netinet/in_pcb.h>
#include <netinet/ip_var.h>
#include <netinet/ip_icmp.h>
#include <sys/socketvar.h>
-#include <sys/kdebug.h>
+#include <netinet/ip_fw.h>
+#include <netinet/ip_divert.h>
+#include <netinet/kpi_ipfilter_var.h>
+
+/* needed for AUTOCONFIGURING: */
+#include <netinet/udp.h>
+#include <netinet/udp_var.h>
+#include <netinet/bootp.h>
+
+#include <sys/kdebug.h>
#define DBG_LAYER_BEG NETDBG_CODE(DBG_NETIP, 0)
#define DBG_LAYER_END NETDBG_CODE(DBG_NETIP, 2)
#define DBG_FNC_IP_INPUT NETDBG_CODE(DBG_NETIP, (2 << 8))
-#if IPFIREWALL
-#include <netinet/ip_fw.h>
-#endif
-
#if IPSEC
#include <netinet6/ipsec.h>
#include <netkey/key.h>
-#include <netkey/key_debug.h>
#endif
#include "faith.h"
#include <netinet/ip_dummynet.h>
#endif
+#if IPSEC
+extern int ipsec_bypass;
+extern lck_mtx_t *sadb_mutex;
+#endif
+
int rsvp_on = 0;
static int ip_rsvp_on;
struct socket *ip_rsvpd;
int ipforwarding = 0;
SYSCTL_INT(_net_inet_ip, IPCTL_FORWARDING, forwarding, CTLFLAG_RW,
- &ipforwarding, 0, "");
+ &ipforwarding, 0, "Enable IP forwarding between interfaces");
static int ipsendredirects = 1; /* XXX */
SYSCTL_INT(_net_inet_ip, IPCTL_SENDREDIRECTS, redirect, CTLFLAG_RW,
- &ipsendredirects, 0, "");
+ &ipsendredirects, 0, "Enable sending IP redirects");
int ip_defttl = IPDEFTTL;
SYSCTL_INT(_net_inet_ip, IPCTL_DEFTTL, ttl, CTLFLAG_RW,
- &ip_defttl, 0, "");
+ &ip_defttl, 0, "Maximum TTL on IP packets");
static int ip_dosourceroute = 0;
SYSCTL_INT(_net_inet_ip, IPCTL_SOURCEROUTE, sourceroute, CTLFLAG_RW,
- &ip_dosourceroute, 0, "");
+ &ip_dosourceroute, 0, "Enable forwarding source routed IP packets");
static int ip_acceptsourceroute = 0;
-SYSCTL_INT(_net_inet_ip, IPCTL_ACCEPTSOURCEROUTE, accept_sourceroute,
- CTLFLAG_RW, &ip_acceptsourceroute, 0, "");
+SYSCTL_INT(_net_inet_ip, IPCTL_ACCEPTSOURCEROUTE, accept_sourceroute,
+ CTLFLAG_RW, &ip_acceptsourceroute, 0,
+ "Enable accepting source routed IP packets");
static int ip_keepfaith = 0;
SYSCTL_INT(_net_inet_ip, IPCTL_KEEPFAITH, keepfaith, CTLFLAG_RW,
- &ip_keepfaith, 0, "");
+ &ip_keepfaith, 0,
+ "Enable packet capture for FAITH IPv4->IPv6 translater daemon");
+
+static int nipq = 0; /* total # of reass queues */
+static int maxnipq;
+SYSCTL_INT(_net_inet_ip, OID_AUTO, maxfragpackets, CTLFLAG_RW,
+ &maxnipq, 0,
+ "Maximum number of IPv4 fragment reassembly queue entries");
+
+static int maxfragsperpacket;
+SYSCTL_INT(_net_inet_ip, OID_AUTO, maxfragsperpacket, CTLFLAG_RW,
+ &maxfragsperpacket, 0,
+ "Maximum number of IPv4 fragments allowed per packet");
+
+static int maxfrags;
+SYSCTL_INT(_net_inet_ip, OID_AUTO, maxfrags, CTLFLAG_RW,
+ &maxfrags, 0, "Maximum number of IPv4 fragments allowed");
+
+static int currentfrags = 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.
+ */
+static int ip_checkinterface = 0;
+SYSCTL_INT(_net_inet_ip, OID_AUTO, check_interface, CTLFLAG_RW,
+ &ip_checkinterface, 0, "Verify packet arrives on correct interface");
#if DIAGNOSTIC
static int ipprintfs = 0;
static int ipqmaxlen = IFQ_MAXLEN;
struct in_ifaddrhead in_ifaddrhead; /* first inet address */
struct ifqueue ipintrq;
-SYSCTL_INT(_net_inet_ip, IPCTL_INTRQMAXLEN, intr_queue_maxlen, CTLFLAG_RD,
- &ipintrq.ifq_maxlen, 0, "");
+SYSCTL_INT(_net_inet_ip, IPCTL_INTRQMAXLEN, intr_queue_maxlen, CTLFLAG_RW,
+ &ipintrq.ifq_maxlen, 0, "Maximum size of the IP input queue");
SYSCTL_INT(_net_inet_ip, IPCTL_INTRQDROPS, intr_queue_drops, CTLFLAG_RD,
- &ipintrq.ifq_drops, 0, "");
+ &ipintrq.ifq_drops, 0, "Number of packets dropped from the IP input queue");
struct ipstat ipstat;
SYSCTL_STRUCT(_net_inet_ip, IPCTL_STATS, stats, CTLFLAG_RD,
- &ipstat, ipstat, "");
+ &ipstat, ipstat, "IP statistics (struct ipstat, netinet/ip_var.h)");
/* Packet reassembly stuff */
#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)
+ (((((x) & 0xF) | ((((x) >> 8) & 0xF) << 4)) ^ (y)) & IPREASS_HMASK)
static struct ipq ipq[IPREASS_NHASH];
-static int nipq = 0; /* total # of reass queues */
-static int maxnipq;
+static TAILQ_HEAD(ipq_list, ipq) ipq_list =
+ TAILQ_HEAD_INITIALIZER(ipq_list);
+const int ipintrq_present = 1;
+lck_mtx_t *ip_mutex;
+lck_attr_t *ip_mutex_attr;
+lck_grp_t *ip_mutex_grp;
+lck_grp_attr_t *ip_mutex_grp_attr;
+lck_mtx_t *inet_domain_mutex;
+extern lck_mtx_t *domain_proto_mtx;
#if IPCTL_DEFMTU
SYSCTL_INT(_net_inet_ip, IPCTL_DEFMTU, mtu, CTLFLAG_RW,
- &ip_mtu, 0, "");
+ &ip_mtu, 0, "Default MTU");
#endif
-#if !defined(COMPAT_IPFW) || COMPAT_IPFW == 1
-#undef COMPAT_IPFW
-#define COMPAT_IPFW 1
-#else
-#undef COMPAT_IPFW
+#if IPSTEALTH
+static int ipstealth = 0;
+SYSCTL_INT(_net_inet_ip, OID_AUTO, stealth, CTLFLAG_RW,
+ &ipstealth, 0, "");
#endif
-#if COMPAT_IPFW
-
-#include <netinet/ip_fw.h>
/* Firewall hooks */
ip_fw_chk_t *ip_fw_chk_ptr;
-ip_fw_ctl_t *ip_fw_ctl_ptr;
+int fw_enable = 1 ;
+int fw_one_pass = 1;
#if DUMMYNET
-ip_dn_ctl_t *ip_dn_ctl_ptr;
+ip_dn_io_t *ip_dn_io_ptr;
#endif
-/* IP Network Address Translation (NAT) hooks */
-ip_nat_t *ip_nat_ptr;
-ip_nat_ctl_t *ip_nat_ctl_ptr;
-#endif
+int (*fr_checkp)(struct ip *, int, struct ifnet *, int, struct mbuf **) = NULL;
+
+SYSCTL_NODE(_net_inet_ip, OID_AUTO, linklocal, CTLFLAG_RW, 0, "link local");
+
+struct ip_linklocal_stat ip_linklocal_stat;
+SYSCTL_STRUCT(_net_inet_ip_linklocal, OID_AUTO, stat, CTLFLAG_RD,
+ &ip_linklocal_stat, ip_linklocal_stat,
+ "Number of link local packets with TTL less than 255");
+
+SYSCTL_NODE(_net_inet_ip_linklocal, OID_AUTO, in, CTLFLAG_RW, 0, "link local input");
+
+int ip_linklocal_in_allowbadttl = 1;
+SYSCTL_INT(_net_inet_ip_linklocal_in, OID_AUTO, allowbadttl, CTLFLAG_RW,
+ &ip_linklocal_in_allowbadttl, 0,
+ "Allow incoming link local packets with TTL less than 255");
+
/*
* We need to save the IP options in case a protocol wants to respond
struct in_addr route[MAX_IPOPTLEN/sizeof(struct in_addr)];
} ip_srcrt;
-#if IPDIVERT
-/*
- * Shared variable between ip_input() and ip_reass() to communicate
- * about which packets, once assembled from fragments, get diverted,
- * and to which port.
- */
-static u_short frag_divert_port;
+#ifdef __APPLE__
+extern struct mbuf* m_dup(register struct mbuf *m, int how);
#endif
-struct sockaddr_in *ip_fw_fwd_addr;
-
-static void save_rte __P((u_char *, struct in_addr));
-static int ip_dooptions __P((struct mbuf *));
-#ifndef NATPT
-static
+static void save_rte(u_char *, struct in_addr);
+static int ip_dooptions(struct mbuf *, int, struct sockaddr_in *, struct route *ipforward_rt);
+static void ip_forward(struct mbuf *, int, struct sockaddr_in *, struct route *ipforward_rt);
+static void ip_freef(struct ipq *);
+#if IPDIVERT
+#ifdef IPDIVERT_44
+static struct mbuf *ip_reass(struct mbuf *,
+ struct ipq *, struct ipq *, u_int32_t *, u_int16_t *);
+#else
+static struct mbuf *ip_reass(struct mbuf *,
+ struct ipq *, struct ipq *, u_int16_t *, u_int16_t *);
#endif
-void ip_forward __P((struct mbuf *, int));
-static void ip_freef __P((struct ipq *));
-static struct ip *
- ip_reass __P((struct mbuf *, struct ipq *, struct ipq *));
-static struct in_ifaddr *
- ip_rtaddr __P((struct in_addr));
-void ipintr __P((void));
-
-#if PM
-extern int doNatFil;
-extern int doRoute;
-
-extern int pm_in __P((struct ifnet *, struct ip *, struct mbuf *));
-extern struct route *pm_route __P((struct mbuf *));
+#else
+static struct mbuf *ip_reass(struct mbuf *, struct ipq *, struct ipq *);
#endif
+void ipintr(void);
-#if defined(NATPT)
-extern int ip6_protocol_tr;
-
-int natpt_in4 __P((struct mbuf *, struct mbuf **));
+#if RANDOM_IP_ID
+extern u_short ip_id;
+#endif
-#endif /* NATPT */
+extern u_long route_generation;
+extern int apple_hwcksum_rx;
/*
* IP initialization: fill in IP protocol switch table.
register struct protosw *pr;
register int i;
static ip_initialized = 0;
+ struct timeval timenow;
+
if (!ip_initialized)
{
TAILQ_INIT(&in_ifaddrhead);
- pr = pffindproto(PF_INET, IPPROTO_RAW, SOCK_RAW);
+ pr = pffindproto_locked(PF_INET, IPPROTO_RAW, SOCK_RAW);
if (pr == 0)
panic("ip_init");
for (i = 0; i < IPPROTO_MAX; i++)
for (i = 0; i < IPREASS_NHASH; i++)
ipq[i].next = ipq[i].prev = &ipq[i];
- maxnipq = nmbclusters/4;
+ maxnipq = nmbclusters / 32;
+ maxfrags = maxnipq * 2;
+ maxfragsperpacket = 128; /* enough for 64k in 512 byte fragments */
- ip_id = time_second & 0xffff;
- ipintrq.ifq_maxlen = ipqmaxlen;
-#if DUMMYNET
- ip_dn_init();
-#endif
-#if IPNAT
- ip_nat_init();
+#if RANDOM_IP_ID
+ getmicrouptime(&timenow);
+ ip_id = timenow.tv_sec & 0xffff;
#endif
+ ipintrq.ifq_maxlen = ipqmaxlen;
+
+ ipf_init();
+
+ ip_mutex_grp_attr = lck_grp_attr_alloc_init();
+ lck_grp_attr_setdefault(ip_mutex_grp_attr);
+
+ ip_mutex_grp = lck_grp_alloc_init("ip", ip_mutex_grp_attr);
+
+ ip_mutex_attr = lck_attr_alloc_init();
+
+ lck_attr_setdefault(ip_mutex_attr);
+
+ if ((ip_mutex = lck_mtx_alloc_init(ip_mutex_grp, ip_mutex_attr)) == NULL) {
+ printf("ip_init: can't alloc ip_mutex\n");
+ return;
+ }
+
ip_initialized = 1;
}
}
+static void
+ip_proto_input(
+ protocol_family_t protocol,
+ mbuf_t packet)
+{
+ ip_input(packet);
+}
+
/* Initialize the PF_INET domain, and add in the pre-defined protos */
void
in_dinit()
extern int in_proto_count;
if (!inetdomain_initted)
- { kprintf("Initing %d protosw entries\n", in_proto_count);
+ {
+ kprintf("Initing %d protosw entries\n", in_proto_count);
dp = &inetdomain;
+ dp->dom_flags = DOM_REENTRANT;
for (i=0, pr = &inetsw[0]; i<in_proto_count; i++, pr++)
net_add_proto(pr, dp);
+ inet_domain_mutex = dp->dom_mtx;
inetdomain_initted = 1;
+
+ lck_mtx_unlock(domain_proto_mtx);
+ proto_register_input(PF_INET, ip_proto_input, NULL);
+ lck_mtx_lock(domain_proto_mtx);
}
}
+__private_extern__ void
+ip_proto_dispatch_in(
+ struct mbuf *m,
+ int hlen,
+ u_int8_t proto,
+ ipfilter_t inject_ipfref)
+{
+ struct ipfilter *filter;
+ int seen = (inject_ipfref == 0);
+ int changed_header = 0;
+ struct ip *ip;
+
+ if (!TAILQ_EMPTY(&ipv4_filters)) {
+ ipf_ref();
+ TAILQ_FOREACH(filter, &ipv4_filters, ipf_link) {
+ if (seen == 0) {
+ if ((struct ipfilter *)inject_ipfref == filter)
+ seen = 1;
+ } else if (filter->ipf_filter.ipf_input) {
+ errno_t result;
+
+ if (changed_header == 0) {
+ changed_header = 1;
+ ip = mtod(m, struct ip *);
+ ip->ip_len = htons(ip->ip_len + hlen);
+ ip->ip_off = htons(ip->ip_off);
+ ip->ip_sum = 0;
+ ip->ip_sum = in_cksum(m, hlen);
+ }
+ result = filter->ipf_filter.ipf_input(
+ filter->ipf_filter.cookie, (mbuf_t*)&m, hlen, proto);
+ if (result == EJUSTRETURN) {
+ ipf_unref();
+ return;
+ }
+ if (result != 0) {
+ ipf_unref();
+ m_freem(m);
+ return;
+ }
+ }
+ }
+ ipf_unref();
+ }
+ /*
+ * If there isn't a specific lock for the protocol
+ * we're about to call, use the generic lock for AF_INET.
+ * otherwise let the protocol deal with its own locking
+ */
+ ip = mtod(m, struct ip *);
+
+ if (changed_header) {
+ ip->ip_len = ntohs(ip->ip_len) - hlen;
+ ip->ip_off = ntohs(ip->ip_off);
+ }
+
+ if (!(ip_protox[ip->ip_p]->pr_flags & PR_PROTOLOCK)) {
+ lck_mtx_lock(inet_domain_mutex);
+ (*ip_protox[ip->ip_p]->pr_input)(m, hlen);
+ lck_mtx_unlock(inet_domain_mutex);
+ }
+ else
+ (*ip_protox[ip->ip_p]->pr_input)(m, hlen);
+
+}
+
+/*
+ * ipforward_rt cleared in in_addroute()
+ * when a new route is successfully created.
+ */
static struct sockaddr_in ipaddr = { sizeof(ipaddr), AF_INET };
-static struct route ipforward_rt;
/*
* Ip input routine. Checksum and byte swap header. If fragmented
{
struct ip *ip;
struct ipq *fp;
- struct in_ifaddr *ia;
- int i, hlen, mff;
+ struct in_ifaddr *ia = NULL;
+ int i, hlen, mff, checkif;
u_short sum;
-#if !IPDIVERT /* dummy variable for the firewall code to play with */
- u_short ip_divert_cookie = 0 ;
-#endif
-#if COMPAT_IPFW
- struct ip_fw_chain *rule = NULL ;
-#endif
+ struct in_addr pkt_dst;
+ u_int32_t div_info = 0; /* packet divert/tee info */
+ struct ip_fw_args args;
+ ipfilter_t inject_filter_ref = 0;
+ struct m_tag *tag;
+ struct route ipforward_rt = { 0 };
+
+ lck_mtx_lock(ip_mutex);
+
+ args.eh = NULL;
+ args.oif = NULL;
+ args.rule = NULL;
+ args.divert_rule = 0; /* divert cookie */
+ args.next_hop = NULL;
+
+ /* Grab info from mtags prepended to the chain */
+#if DUMMYNET
+ if ((tag = m_tag_locate(m, KERNEL_MODULE_TAG_ID, KERNEL_TAG_TYPE_DUMMYNET, NULL)) != NULL) {
+ struct dn_pkt_tag *dn_tag;
+
+ dn_tag = (struct dn_pkt_tag *)(tag+1);
+ args.rule = dn_tag->rule;
+
+ m_tag_delete(m, tag);
+ }
+#endif /* DUMMYNET */
-#if IPFIREWALL && DUMMYNET
- /*
- * dummynet packet are prepended a vestigial mbuf with
- * m_type = MT_DUMMYNET and m_data pointing to the matching
- * rule.
- */
- if (m->m_type == MT_DUMMYNET) {
- struct mbuf *m0 = m ;
- rule = (struct ip_fw_chain *)(m->m_data) ;
- m = m->m_next ;
- FREE(m0, M_IPFW);
- ip = mtod(m, struct ip *);
- hlen = IP_VHL_HL(ip->ip_vhl) << 2;
- goto iphack ;
- } else
- rule = NULL ;
-#endif
+ if ((tag = m_tag_locate(m, KERNEL_MODULE_TAG_ID, KERNEL_TAG_TYPE_DIVERT, NULL)) != NULL) {
+ struct divert_tag *div_tag;
+
+ div_tag = (struct divert_tag *)(tag+1);
+ args.divert_rule = div_tag->cookie;
+ m_tag_delete(m, tag);
+ }
+ if ((tag = m_tag_locate(m, KERNEL_MODULE_TAG_ID, KERNEL_TAG_TYPE_IPFORWARD, NULL)) != NULL) {
+ struct ip_fwd_tag *ipfwd_tag;
+
+ ipfwd_tag = (struct ip_fwd_tag *)(tag+1);
+ args.next_hop = ipfwd_tag->next_hop;
+
+ m_tag_delete(m, tag);
+ }
+
#if DIAGNOSTIC
if (m == NULL || (m->m_flags & M_PKTHDR) == 0)
panic("ip_input no HDR");
#endif
+
+ if (args.rule) { /* dummynet already filtered us */
+ ip = mtod(m, struct ip *);
+ hlen = IP_VHL_HL(ip->ip_vhl) << 2;
+ inject_filter_ref = ipf_get_inject_filter(m);
+ goto iphack ;
+ }
+
/*
- * If no IP addresses have been set yet but the interfaces
- * are receiving, can't do anything with incoming packets yet.
- * XXX This is broken! We should be able to receive broadcasts
- * and multicasts even without any local addresses configured.
+ * No need to proccess packet twice if we've
+ * already seen it
*/
- if (TAILQ_EMPTY(&in_ifaddrhead))
- goto bad;
+ inject_filter_ref = ipf_get_inject_filter(m);
+ if (inject_filter_ref != 0) {
+ lck_mtx_unlock(ip_mutex);
+ ip = mtod(m, struct ip *);
+ hlen = IP_VHL_HL(ip->ip_vhl) << 2;
+ ip->ip_len = ntohs(ip->ip_len) - hlen;
+ ip->ip_off = ntohs(ip->ip_off);
+ ip_proto_dispatch_in(m, hlen, ip->ip_p, inject_filter_ref);
+ return;
+ }
+
ipstat.ips_total++;
if (m->m_pkthdr.len < sizeof(struct ip))
if (m->m_len < sizeof (struct ip) &&
(m = m_pullup(m, sizeof (struct ip))) == 0) {
ipstat.ips_toosmall++;
+ lck_mtx_unlock(ip_mutex);
return;
}
ip = mtod(m, struct ip *);
if (hlen > m->m_len) {
if ((m = m_pullup(m, hlen)) == 0) {
ipstat.ips_badhlen++;
+ lck_mtx_unlock(ip_mutex);
return;
}
ip = mtod(m, struct ip *);
}
- if (m->m_pkthdr.rcvif->if_hwassist == 0)
- m->m_pkthdr.csum_flags = 0;
+ /* 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;
+ }
+ }
- if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) {
- sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID);
- } else
- sum = in_cksum(m, hlen);
+ /* IPv4 Link-Local Addresses as defined in <draft-ietf-zeroconf-ipv4-linklocal-05.txt> */
+ if ((IN_LINKLOCAL(ntohl(ip->ip_dst.s_addr)) ||
+ IN_LINKLOCAL(ntohl(ip->ip_src.s_addr)))) {
+ ip_linklocal_stat.iplls_in_total++;
+ if (ip->ip_ttl != MAXTTL) {
+ ip_linklocal_stat.iplls_in_badttl++;
+ /* Silently drop link local traffic with bad TTL */
+ if (!ip_linklocal_in_allowbadttl)
+ goto bad;
+ }
+ }
+ if ((IF_HWASSIST_CSUM_FLAGS(m->m_pkthdr.rcvif->if_hwassist) == 0)
+ || (apple_hwcksum_rx == 0) ||
+ ((m->m_pkthdr.csum_flags & CSUM_TCP_SUM16) && ip->ip_p != IPPROTO_TCP)) {
+ m->m_pkthdr.csum_flags = 0; /* invalidate HW generated checksum flags */
+ }
+ if (m->m_pkthdr.csum_flags & CSUM_IP_CHECKED) {
+ sum = !(m->m_pkthdr.csum_flags & CSUM_IP_VALID);
+ } else {
+ sum = in_cksum(m, hlen);
+ }
if (sum) {
ipstat.ips_badsum++;
goto bad;
ipstat.ips_badlen++;
goto bad;
}
- NTOHS(ip->ip_id);
NTOHS(ip->ip_off);
/*
goto bad;
}
if (m->m_pkthdr.len > ip->ip_len) {
+ /* Invalidate hwcksuming */
+ m->m_pkthdr.csum_flags = 0;
+ m->m_pkthdr.csum_data = 0;
+
if (m->m_len == m->m_pkthdr.len) {
m->m_len = ip->ip_len;
m->m_pkthdr.len = ip->ip_len;
} else
m_adj(m, ip->ip_len - m->m_pkthdr.len);
}
+
+#if IPSEC
+ if (ipsec_bypass == 0 && ipsec_gethist(m, NULL))
+ goto pass;
+#endif
+
/*
* IpHack's section.
* Right now when no processing on packet has done
#if defined(IPFIREWALL) && defined(DUMMYNET)
iphack:
#endif
-#if COMPAT_IPFW
- if (ip_fw_chk_ptr) {
+ /*
+ * Check if we want to allow this packet to be processed.
+ * Consider it to be bad if not.
+ */
+ if (fr_checkp) {
+ struct mbuf *m1 = m;
+
+ if (fr_checkp(ip, hlen, m->m_pkthdr.rcvif, 0, &m1) || !m1) {
+ lck_mtx_unlock(ip_mutex);
+ return;
+ }
+ ip = mtod(m = m1, struct ip *);
+ }
+ if (fw_enable && IPFW_LOADED) {
#if IPFIREWALL_FORWARD
/*
* If we've been forwarded from the output side, then
* skip the firewall a second time
*/
- if (ip_fw_fwd_addr)
+ if (args.next_hop)
goto ours;
#endif /* IPFIREWALL_FORWARD */
- i = (*ip_fw_chk_ptr)(&ip, hlen, NULL, &ip_divert_cookie,
- &m, &rule, &ip_fw_fwd_addr);
- /*
- * see the comment in ip_output for the return values
- * produced by the firewall.
- */
- if (!m) /* packet discarded by firewall */
- return ;
- if (i == 0 && ip_fw_fwd_addr == NULL) /* common case */
- goto pass ;
+
+ args.m = m;
+ lck_mtx_unlock(ip_mutex);
+
+ i = ip_fw_chk_ptr(&args);
+ m = args.m;
+
+ if ( (i & IP_FW_PORT_DENY_FLAG) || m == NULL) { /* drop */
+ if (m)
+ m_freem(m);
+ return;
+ }
+ ip = mtod(m, struct ip *); /* just in case m changed */
+ if (i == 0 && args.next_hop == NULL) { /* common case */
+ lck_mtx_lock(ip_mutex);
+ goto pass;
+ }
#if DUMMYNET
- if (i & 0x10000) {
- /* send packet to the appropriate pipe */
- dummynet_io(i&0xffff,DN_TO_IP_IN,m,NULL,NULL,0, rule);
- return ;
+ if (DUMMYNET_LOADED && (i & IP_FW_PORT_DYNT_FLAG) != 0) {
+ /* Send packet to the appropriate pipe */
+ ip_dn_io_ptr(m, i&0xffff, DN_TO_IP_IN, &args);
+ return;
}
-#endif
+#endif /* DUMMYNET */
#if IPDIVERT
- if (i > 0 && i < 0x10000) {
- /* Divert packet */
- frag_divert_port = i & 0xffff ;
+ if (i != 0 && (i & IP_FW_PORT_DYNT_FLAG) == 0) {
+ /* Divert or tee packet */
+ lck_mtx_lock(ip_mutex);
+ div_info = i;
goto ours;
}
#endif
#if IPFIREWALL_FORWARD
- if (i == 0 && ip_fw_fwd_addr != NULL)
- goto pass ;
+ if (i == 0 && args.next_hop != NULL) {
+ lck_mtx_lock(ip_mutex);
+ goto pass;
+ }
#endif
/*
* if we get here, the packet must be dropped
*/
- m_freem(m);
- return;
- }
-pass:
-
- if (ip_nat_ptr && !(*ip_nat_ptr)(&ip, &m, m->m_pkthdr.rcvif, IP_NAT_IN)) {
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL;
-#endif
- return;
- }
-#endif /* !COMPAT_IPFW */
-
-#if defined(PM)
- /*
- * Process ip-filter/NAT.
- * Return TRUE if this packed is discarded.
- * Return FALSE if this packed is accepted.
- */
-
- if (doNatFil && pm_in(m->m_pkthdr.rcvif, ip, m))
- return;
-#endif
-
-#if defined(NATPT)
- /*
- *
- */
- if (ip6_protocol_tr)
- {
- struct mbuf *m1 = NULL;
-
- switch (natpt_in4(m, &m1))
- {
- case IPPROTO_IP: goto dooptions;
- case IPPROTO_IPV4: ip_forward(m1, 0); break;
- case IPPROTO_IPV6: ip6_forward(m1, 1); break;
- case IPPROTO_MAX: /* discard this packet */
- default:
- }
-
- if (m != m1)
m_freem(m);
-
- return;
+ return;
}
- dooptions:
-#endif
+pass:
/*
* Process options and, if not destined for us,
* to be sent and the original packet to be freed).
*/
ip_nhops = 0; /* for source routed packets */
- if (hlen > sizeof (struct ip) && ip_dooptions(m)) {
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL;
-#endif
+ if (hlen > sizeof (struct ip) && ip_dooptions(m, 0, args.next_hop, &ipforward_rt)) {
+ lck_mtx_unlock(ip_mutex);
return;
}
/*
* Check our list of addresses, to see if the packet is for us.
+ * If we don't have any addresses, assume any unicast packet
+ * we receive might be for us (and let the upper layers deal
+ * with it).
*/
- for (ia = TAILQ_FIRST(&in_ifaddrhead); ia;
- ia = TAILQ_NEXT(ia, ia_link)) {
-#define satosin(sa) ((struct sockaddr_in *)(sa))
+ if (TAILQ_EMPTY(&in_ifaddrhead) &&
+ (m->m_flags & (M_MCAST|M_BCAST)) == 0)
+ goto ours;
- if (IA_SIN(ia)->sin_addr.s_addr == ip->ip_dst.s_addr)
- goto ours;
+ /*
+ * Cache the destination address of the packet; this may be
+ * changed by use of 'ipfw fwd'.
+ */
+ pkt_dst = args.next_hop == NULL ?
+ ip->ip_dst : args.next_hop->sin_addr;
- if (IA_SIN(ia)->sin_addr.s_addr == INADDR_ANY)
- goto ours;
+ /*
+ * 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 and the packet is not subject to
+ * 'ipfw fwd'.
+ *
+ * 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.
+ */
+ checkif = ip_checkinterface && (ipforwarding == 0) &&
+ ((m->m_pkthdr.rcvif->if_flags & IFF_LOOPBACK) == 0) &&
+ (args.next_hop == NULL);
-#if IPFIREWALL_FORWARD
+ lck_mtx_lock(rt_mtx);
+ TAILQ_FOREACH(ia, &in_ifaddrhead, ia_link) {
+#define satosin(sa) ((struct sockaddr_in *)(sa))
+
+ if (IA_SIN(ia)->sin_addr.s_addr == INADDR_ANY) {
+ lck_mtx_unlock(rt_mtx);
+ goto ours;
+ }
+
/*
- * If the addr to forward to is one of ours, we pretend to
- * be the destination for this packet.
+ * If the address matches, verify that the packet
+ * arrived via the correct interface if checking is
+ * enabled.
*/
- if (ip_fw_fwd_addr == NULL) {
- if (IA_SIN(ia)->sin_addr.s_addr == ip->ip_dst.s_addr)
- goto ours;
- } else if (IA_SIN(ia)->sin_addr.s_addr ==
- ip_fw_fwd_addr->sin_addr.s_addr)
- goto ours;
-#else
- if (IA_SIN(ia)->sin_addr.s_addr == ip->ip_dst.s_addr)
+ if (IA_SIN(ia)->sin_addr.s_addr == pkt_dst.s_addr &&
+ (!checkif || ia->ia_ifp == m->m_pkthdr.rcvif)) {
+ lck_mtx_unlock(rt_mtx);
goto ours;
-#endif
- if (ia->ia_ifp && ia->ia_ifp->if_flags & IFF_BROADCAST) {
+ }
+ /*
+ * Only accept broadcast packets that arrive via the
+ * matching interface. Reception of forwarded directed
+ * broadcasts would be handled via ip_forward() and
+ * ether_output() with the loopback into the stack for
+ * SIMPLEX interfaces handled by ether_output().
+ */
+ if ((!checkif || ia->ia_ifp == m->m_pkthdr.rcvif) &&
+ ia->ia_ifp && ia->ia_ifp->if_flags & IFF_BROADCAST) {
if (satosin(&ia->ia_broadaddr)->sin_addr.s_addr ==
- ip->ip_dst.s_addr)
+ pkt_dst.s_addr) {
+ lck_mtx_unlock(rt_mtx);
goto ours;
- if (ip->ip_dst.s_addr == ia->ia_netbroadcast.s_addr)
+ }
+ if (ia->ia_netbroadcast.s_addr == pkt_dst.s_addr) {
+ lck_mtx_unlock(rt_mtx);
goto ours;
+ }
}
}
+ lck_mtx_unlock(rt_mtx);
if (IN_MULTICAST(ntohl(ip->ip_dst.s_addr))) {
struct in_multi *inm;
if (ip_mrouter) {
* The packet is returned (relatively) intact; if
* ip_mforward() returns a non-zero value, the packet
* must be discarded, else it may be accepted below.
- *
- * (The IP ident field is put in the same byte order
- * as expected when ip_mforward() is called from
- * ip_output().)
*/
- ip->ip_id = htons(ip->ip_id);
- if (ip_mforward(ip, m->m_pkthdr.rcvif, m, 0) != 0) {
+ if (ip_mforward &&
+ ip_mforward(ip, m->m_pkthdr.rcvif, m, 0) != 0) {
ipstat.ips_cantforward++;
m_freem(m);
+ lck_mtx_unlock(ip_mutex);
return;
}
- ip->ip_id = ntohs(ip->ip_id);
/*
- * The process-level routing demon needs to receive
+ * The process-level routing daemon needs to receive
* all multicast IGMP packets, whether or not this
* host belongs to their destination groups.
*/
if (inm == NULL) {
ipstat.ips_notmember++;
m_freem(m);
+ lck_mtx_unlock(ip_mutex);
return;
}
goto ours;
if (ip->ip_dst.s_addr == INADDR_ANY)
goto ours;
- if (m->m_pkthdr.rcvif
+ /* Allow DHCP/BootP responses through */
+ if (m->m_pkthdr.rcvif != NULL
&& (m->m_pkthdr.rcvif->if_eflags & IFEF_AUTOCONFIGURING)
+ && hlen == sizeof(struct ip)
&& ip->ip_p == IPPROTO_UDP) {
- goto ours;
+ struct udpiphdr *ui;
+ if (m->m_len < sizeof(struct udpiphdr)
+ && (m = m_pullup(m, sizeof(struct udpiphdr))) == 0) {
+ udpstat.udps_hdrops++;
+ lck_mtx_unlock(ip_mutex);
+ return;
+ }
+ ui = mtod(m, struct udpiphdr *);
+ if (ntohs(ui->ui_dport) == IPPORT_BOOTPC) {
+ goto ours;
+ }
+ ip = mtod(m, struct ip *); /* in case it changed */
}
-#if defined(NFAITH) && NFAITH > 0
+#if defined(NFAITH) && 0 < NFAITH
/*
* FAITH(Firewall Aided Internet Translator)
*/
goto ours;
}
m_freem(m);
+ lck_mtx_unlock(ip_mutex);
return;
}
#endif
+ lck_mtx_unlock(ip_mutex);
/*
* Not for us; forward if possible and desirable.
*/
if (ipforwarding == 0) {
ipstat.ips_cantforward++;
m_freem(m);
- } else
- ip_forward(m, 0);
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL;
-#endif
+ } else {
+ ip_forward(m, 0, args.next_hop, &ipforward_rt);
+ }
return;
ours:
+#ifndef __APPLE__
+ /* Darwin does not have an if_data in ifaddr */
+ /* Count the packet in the ip address stats */
+ if (ia != NULL) {
+ ia->ia_ifa.if_ipackets++;
+ ia->ia_ifa.if_ibytes += m->m_pkthdr.len;
+ }
+#endif
/*
* If offset or IP_MF are set, must reassemble.
* but it's not worth the time; just let them time out.)
*/
if (ip->ip_off & (IP_MF | IP_OFFMASK | IP_RF)) {
- if (m->m_flags & M_EXT) { /* XXX */
- if ((m = m_pullup(m, hlen)) == 0) {
- ipstat.ips_toosmall++;
-#if IPDIVERT
- frag_divert_port = 0;
- ip_divert_cookie = 0;
-#endif
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL;
-#endif
- return;
+
+ /* If maxnipq is 0, never accept fragments. */
+ if (maxnipq == 0) {
+ ipstat.ips_fragments++;
+ ipstat.ips_fragdropped++;
+ goto bad;
+ }
+
+ /*
+ * If we will exceed the number of fragments in queues, timeout the
+ * oldest fragemented packet to make space.
+ */
+ if (currentfrags >= maxfrags) {
+ fp = TAILQ_LAST(&ipq_list, ipq_list);
+ ipstat.ips_fragtimeout += fp->ipq_nfrags;
+
+ if (ip->ip_id == fp->ipq_id &&
+ ip->ip_src.s_addr == fp->ipq_src.s_addr &&
+ ip->ip_dst.s_addr == fp->ipq_dst.s_addr &&
+ ip->ip_p == fp->ipq_p) {
+ /*
+ * If we match the fragment queue we were going to
+ * discard, drop this packet too.
+ */
+ ipstat.ips_fragdropped++;
+ ip_freef(fp);
+ goto bad;
}
- ip = mtod(m, struct ip *);
+
+ ip_freef(fp);
}
+
sum = IPREASS_HASH(ip->ip_src.s_addr, ip->ip_id);
/*
* Look for queue of fragments
ip->ip_p == fp->ipq_p)
goto found;
- fp = 0;
-
- /* check if there's a place for the new queue */
- if (nipq > maxnipq) {
+ /*
+ * Enforce upper bound on number of fragmented packets
+ * for which we attempt reassembly;
+ * If maxnipq is -1, accept all fragments without limitation.
+ */
+ if ((nipq > maxnipq) && (maxnipq > 0)) {
/*
- * drop something from the tail of the current queue
- * before proceeding further
+ * drop the oldest fragment before proceeding further
*/
- if (ipq[sum].prev == &ipq[sum]) { /* gak */
- for (i = 0; i < IPREASS_NHASH; i++) {
- if (ipq[i].prev != &ipq[i]) {
- ip_freef(ipq[i].prev);
- break;
- }
- }
- } else
- ip_freef(ipq[sum].prev);
+ fp = TAILQ_LAST(&ipq_list, ipq_list);
+ ipstat.ips_fragtimeout += fp->ipq_nfrags;
+ ip_freef(fp);
}
+
+ fp = NULL;
+
found:
/*
* Adjust ip_len to not reflect header,
- * set ip_mff if more fragments are expected,
* convert offset of this to bytes.
*/
ip->ip_len -= hlen;
- mff = (ip->ip_off & IP_MF) != 0;
- if (mff) {
+ if (ip->ip_off & IP_MF) {
/*
* 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) {
ipstat.ips_toosmall++; /* XXX */
goto bad;
}
m->m_flags |= M_FRAG;
+ } else {
+ /* Clear the flag in case packet comes from loopback */
+ m->m_flags &= ~M_FRAG;
}
ip->ip_off <<= 3;
/*
- * If datagram marked as having more fragments
- * or if this is not the first fragment,
- * attempt reassembly; if it succeeds, proceed.
+ * Attempt reassembly; if it succeeds, proceed.
+ * ip_reass() will return a different mbuf, and update
+ * the divert info in div_info and args.divert_rule.
*/
- if (mff || ip->ip_off) {
ipstat.ips_fragments++;
m->m_pkthdr.header = ip;
- ip = ip_reass(m, fp, &ipq[sum]);
- if (ip == 0) {
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL;
+#if IPDIVERT
+ m = ip_reass(m,
+ fp, &ipq[sum], &div_info, &args.divert_rule);
+#else
+ m = ip_reass(m, fp, &ipq[sum]);
#endif
+ if (m == 0) {
+ lck_mtx_unlock(ip_mutex);
return;
}
- /* Get the length of the reassembled packets header */
- hlen = IP_VHL_HL(ip->ip_vhl) << 2;
ipstat.ips_reassembled++;
- m = dtom(ip);
+ ip = mtod(m, struct ip *);
+ /* Get the header length of the reassembled packet */
+ hlen = IP_VHL_HL(ip->ip_vhl) << 2;
#if IPDIVERT
- if (frag_divert_port) {
- struct mbuf m;
- m.m_next = 0;
- m.m_len = hlen;
- m.m_data = (char *) ip;
+ /* Restore original checksum before diverting packet */
+ if (div_info != 0) {
ip->ip_len += hlen;
HTONS(ip->ip_len);
HTONS(ip->ip_off);
- HTONS(ip->ip_id);
ip->ip_sum = 0;
- ip->ip_sum = in_cksum(&m, hlen);
- NTOHS(ip->ip_id);
+ ip->ip_sum = in_cksum(m, hlen);
NTOHS(ip->ip_off);
NTOHS(ip->ip_len);
ip->ip_len -= hlen;
}
#endif
} else
- if (fp)
- ip_freef(fp);
- } else
ip->ip_len -= hlen;
#if IPDIVERT
/*
- * Divert reassembled packets to the divert protocol if required
- * If divert port is null then cookie should be too,
- * so we shouldn't need to clear them here. Assume ip_divert does so.
+ * Divert or tee packet to the divert protocol if required.
+ *
+ * If div_info is zero then cookie should be too, so we shouldn't
+ * need to clear them here. Assume divert_packet() does so also.
*/
- if (frag_divert_port) {
+ if (div_info != 0) {
+ struct mbuf *clone = NULL;
+
+ /* Clone packet if we're doing a 'tee' */
+ if ((div_info & IP_FW_PORT_TEE_FLAG) != 0)
+ clone = m_dup(m, M_DONTWAIT);
+
+ /* Restore packet header fields to original values */
+ ip->ip_len += hlen;
+ HTONS(ip->ip_len);
+ HTONS(ip->ip_off);
+
+ /* Deliver packet to divert input routine */
ipstat.ips_delivered++;
- ip_divert_port = frag_divert_port;
- frag_divert_port = 0;
- (*ip_protox[IPPROTO_DIVERT]->pr_input)(m, hlen);
- return;
- }
+ lck_mtx_unlock(ip_mutex);
+ divert_packet(m, 1, div_info & 0xffff, args.divert_rule);
- /* Don't let packets divert themselves */
- if (ip->ip_p == IPPROTO_DIVERT) {
- ipstat.ips_noproto++;
- goto bad;
+ /* If 'tee', continue with original packet */
+ if (clone == NULL) {
+ return;
+ }
+ lck_mtx_lock(ip_mutex);
+ m = clone;
+ ip = mtod(m, struct ip *);
}
+#endif
+#if 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 (ipsec_bypass == 0 && (ip_protox[ip->ip_p]->pr_flags & PR_LASTHDR) != 0) {
+ lck_mtx_lock(sadb_mutex);
+ if (ipsec4_in_reject(m, NULL)) {
+ ipsecstat.in_polvio++;
+ lck_mtx_unlock(sadb_mutex);
+ goto bad;
+ }
+ lck_mtx_unlock(sadb_mutex);
+ }
#endif
/*
* Switch out to protocol's input routine.
*/
ipstat.ips_delivered++;
-
- KERNEL_DEBUG(DBG_LAYER_END, ip->ip_dst.s_addr,
- ip->ip_src.s_addr, ip->ip_p, ip->ip_off, ip->ip_len);
-
- (*ip_protox[ip->ip_p]->pr_input)(m, hlen);
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL; /* tcp needed it */
-#endif
- return;
+ {
+ if (args.next_hop && ip->ip_p == IPPROTO_TCP) {
+ /* TCP needs IPFORWARD info if available */
+ struct m_tag *fwd_tag;
+ struct ip_fwd_tag *ipfwd_tag;
+
+ fwd_tag = m_tag_alloc(KERNEL_MODULE_TAG_ID, KERNEL_TAG_TYPE_IPFORWARD,
+ sizeof(struct sockaddr_in), M_NOWAIT);
+ if (fwd_tag == NULL) {
+ goto bad;
+ }
+
+ ipfwd_tag = (struct ip_fwd_tag *)(fwd_tag+1);
+ ipfwd_tag->next_hop = args.next_hop;
+
+ m_tag_prepend(m, fwd_tag);
+
+ KERNEL_DEBUG(DBG_LAYER_END, ip->ip_dst.s_addr,
+ ip->ip_src.s_addr, ip->ip_p, ip->ip_off, ip->ip_len);
+
+ lck_mtx_unlock(ip_mutex);
+
+ /* TCP deals with its own locking */
+ ip_proto_dispatch_in(m, hlen, ip->ip_p, 0);
+ } else {
+ KERNEL_DEBUG(DBG_LAYER_END, ip->ip_dst.s_addr,
+ ip->ip_src.s_addr, ip->ip_p, ip->ip_off, ip->ip_len);
+
+ lck_mtx_unlock(ip_mutex);
+ ip_proto_dispatch_in(m, hlen, ip->ip_p, 0);
+ }
+
+ return;
+ }
bad:
-#if IPFIREWALL_FORWARD
- ip_fw_fwd_addr = NULL;
-#endif
KERNEL_DEBUG(DBG_LAYER_END, 0,0,0,0,0);
+ lck_mtx_unlock(ip_mutex);
m_freem(m);
}
/*
- * IP software interrupt routine - to go away sometime soon
+ * Take incoming datagram fragment and try to reassemble it into
+ * whole datagram. If a chain for reassembly of this datagram already
+ * exists, then it is given as fp; otherwise have to make a chain.
+ *
+ * When IPDIVERT enabled, keep additional state with each packet that
+ * tells us if we need to divert or tee the packet we're building.
*/
-void
-ipintr(void)
-{
- int s;
- struct mbuf *m;
-
- KERNEL_DEBUG(DBG_FNC_IP_INPUT | DBG_FUNC_START, 0,0,0,0,0);
-
- while(1) {
- s = splimp();
- IF_DEQUEUE(&ipintrq, m);
- splx(s);
- if (m == 0) {
- KERNEL_DEBUG(DBG_FNC_IP_INPUT | DBG_FUNC_END, 0,0,0,0,0);
- return;
- }
-
- ip_input(m);
- }
-}
-NETISR_SET(NETISR_IP, ipintr);
-
-/*
- * Take incoming datagram fragment and try to
- * reassemble it into whole datagram. If a chain for
- * reassembly of this datagram already exists, then it
- * is given as fp; otherwise have to make a chain.
- */
-static struct ip *
+static struct mbuf *
+#if IPDIVERT
+ip_reass(m, fp, where, divinfo, divcookie)
+#else
ip_reass(m, fp, where)
+#endif
register struct mbuf *m;
register struct ipq *fp;
struct ipq *where;
+#if IPDIVERT
+#ifdef IPDIVERT_44
+ u_int32_t *divinfo;
+#else
+ u_int16_t *divinfo;
+#endif
+ u_int16_t *divcookie;
+#endif
{
struct ip *ip = mtod(m, struct ip *);
register struct mbuf *p = 0, *q, *nq;
if ((t = m_get(M_DONTWAIT, MT_FTABLE)) == NULL)
goto dropfrag;
fp = mtod(t, struct ipq *);
- insque((void *) fp, (void *) where);
+ insque((void*)fp, (void*)where);
nipq++;
+ fp->ipq_nfrags = 1;
fp->ipq_ttl = IPFRAGTTL;
fp->ipq_p = ip->ip_p;
fp->ipq_id = ip->ip_id;
fp->ipq_frags = m;
m->m_nextpkt = NULL;
#if IPDIVERT
+#ifdef IPDIVERT_44
+ fp->ipq_div_info = 0;
+#else
fp->ipq_divert = 0;
+#endif
fp->ipq_div_cookie = 0;
#endif
+ TAILQ_INSERT_HEAD(&ipq_list, fp, ipq_list);
goto inserted;
+ } else {
+ fp->ipq_nfrags++;
}
#define GETIP(m) ((struct ip*)((m)->m_pkthdr.header))
* our data already. If so, drop the data from the incoming
* segment. If it provides all of our data, drop us, otherwise
* stick new segment in the proper place.
+ *
+ * If some of the data is dropped from the the preceding
+ * segment, then it's checksum is invalidated.
*/
if (p) {
i = GETIP(p)->ip_off + GETIP(p)->ip_len - ip->ip_off;
if (i > 0) {
if (i >= ip->ip_len)
goto dropfrag;
- m_adj(dtom(ip), i);
- m->m_pkthdr.csum_flags = 0;
+ m_adj(m, i);
+ m->m_pkthdr.csum_flags = 0;
ip->ip_off += i;
ip->ip_len -= i;
}
GETIP(q)->ip_len -= i;
GETIP(q)->ip_off += i;
m_adj(q, i);
- q->m_pkthdr.csum_flags = 0;
+ q->m_pkthdr.csum_flags = 0;
break;
}
nq = q->m_nextpkt;
m->m_nextpkt = nq;
+ ipstat.ips_fragdropped++;
+ fp->ipq_nfrags--;
m_freem(q);
}
inserted:
+ currentfrags++;
#if IPDIVERT
/*
- * Any fragment diverting causes the whole packet to divert
+ * Transfer firewall instructions to the fragment structure.
+ * Only trust info in the fragment at offset 0.
*/
- if (frag_divert_port) {
- fp->ipq_divert = frag_divert_port;
- fp->ipq_div_cookie = ip_divert_cookie;
+ if (ip->ip_off == 0) {
+#ifdef IPDIVERT_44
+ fp->ipq_div_info = *divinfo;
+#else
+ fp->ipq_divert = *divinfo;
+#endif
+ fp->ipq_div_cookie = *divcookie;
}
- frag_divert_port = 0;
- ip_divert_cookie = 0;
+ *divinfo = 0;
+ *divcookie = 0;
#endif
/*
- * Check for complete reassembly.
+ * Check for complete reassembly and perform frag per packet
+ * limiting.
+ *
+ * Frag limiting is performed here so that the nth frag has
+ * a chance to complete the packet before we drop the packet.
+ * As a result, n+1 frags are actually allowed per packet, but
+ * only n will ever be stored. (n = maxfragsperpacket.)
+ *
*/
next = 0;
for (p = NULL, q = fp->ipq_frags; q; p = q, q = q->m_nextpkt) {
- if (GETIP(q)->ip_off != next)
+ if (GETIP(q)->ip_off != next) {
+ if (fp->ipq_nfrags > maxfragsperpacket) {
+ ipstat.ips_fragdropped += fp->ipq_nfrags;
+ ip_freef(fp);
+ }
return (0);
+ }
next += GETIP(q)->ip_len;
}
/* Make sure the last packet didn't have the IP_MF flag */
- if (p->m_flags & M_FRAG)
+ if (p->m_flags & M_FRAG) {
+ if (fp->ipq_nfrags > maxfragsperpacket) {
+ ipstat.ips_fragdropped += fp->ipq_nfrags;
+ ip_freef(fp);
+ }
return (0);
+ }
/*
* Reassembly is complete. Make sure the packet is a sane size.
ip = GETIP(q);
if (next + (IP_VHL_HL(ip->ip_vhl) << 2) > IP_MAXPACKET) {
ipstat.ips_toolong++;
+ ipstat.ips_fragdropped += fp->ipq_nfrags;
ip_freef(fp);
return (0);
}
nq = q->m_nextpkt;
q->m_nextpkt = NULL;
if (q->m_pkthdr.csum_flags & CSUM_TCP_SUM16)
- m->m_pkthdr.csum_flags = 0;
+ m->m_pkthdr.csum_flags = 0;
else {
- m->m_pkthdr.csum_data += q->m_pkthdr.csum_data ;
- m->m_pkthdr.csum_flags &= q->m_pkthdr.csum_flags;
+ m->m_pkthdr.csum_flags &= q->m_pkthdr.csum_flags;
+ m->m_pkthdr.csum_data += q->m_pkthdr.csum_data;
}
m_cat(m, q);
}
#if IPDIVERT
/*
- * extract divert port for packet, if any
+ * Extract firewall instructions from the fragment structure.
*/
- frag_divert_port = fp->ipq_divert;
- ip_divert_cookie = fp->ipq_div_cookie;
+#ifdef IPDIVERT_44
+ *divinfo = fp->ipq_div_info;
+#else
+ *divinfo = fp->ipq_divert;
+#endif
+ *divcookie = fp->ipq_div_cookie;
#endif
/*
ip->ip_len = next;
ip->ip_src = fp->ipq_src;
ip->ip_dst = fp->ipq_dst;
- remque((void *) fp);
+ remque((void*)fp);
+ TAILQ_REMOVE(&ipq_list, fp, ipq_list);
+ currentfrags -= fp->ipq_nfrags;
nipq--;
(void) m_free(dtom(fp));
m->m_len += (IP_VHL_HL(ip->ip_vhl) << 2);
/* some debugging cruft by sklower, below, will go away soon */
if (m->m_flags & M_PKTHDR) { /* XXX this should be done elsewhere */
register int plen = 0;
- for (t = m; m; m = m->m_next)
- plen += m->m_len;
- t->m_pkthdr.len = plen;
+ for (t = m; t; t = t->m_next)
+ plen += t->m_len;
+ m->m_pkthdr.len = plen;
}
- return (ip);
+ return (m);
dropfrag:
#if IPDIVERT
- frag_divert_port = 0;
- ip_divert_cookie = 0;
+ *divinfo = 0;
+ *divcookie = 0;
#endif
ipstat.ips_fragdropped++;
+ if (fp != 0)
+ fp->ipq_nfrags--;
m_freem(m);
return (0);
ip_freef(fp)
struct ipq *fp;
{
- register struct mbuf *q;
-
- while (fp->ipq_frags) {
- q = fp->ipq_frags;
- fp->ipq_frags = q->m_nextpkt;
- m_freem(q);
- }
- remque((void *) fp);
+ currentfrags -= fp->ipq_nfrags;
+ m_freem_list(fp->ipq_frags);
+ remque((void*)fp);
+ TAILQ_REMOVE(&ipq_list, fp, ipq_list);
(void) m_free(dtom(fp));
nipq--;
}
ip_slowtimo()
{
register struct ipq *fp;
- int s = splnet();
int i;
-
+ lck_mtx_lock(ip_mutex);
for (i = 0; i < IPREASS_NHASH; i++) {
fp = ipq[i].next;
if (fp == 0)
--fp->ipq_ttl;
fp = fp->next;
if (fp->prev->ipq_ttl == 0) {
- ipstat.ips_fragtimeout++;
+ ipstat.ips_fragtimeout += fp->prev->ipq_nfrags;
ip_freef(fp->prev);
}
}
}
+ /*
+ * If we are over the maximum number of fragments
+ * (due to the limit being lowered), drain off
+ * enough to get down to the new limit.
+ */
+ if (maxnipq >= 0 && nipq > maxnipq) {
+ for (i = 0; i < IPREASS_NHASH; i++) {
+ while (nipq > maxnipq &&
+ (ipq[i].next != &ipq[i])) {
+ ipstat.ips_fragdropped +=
+ ipq[i].next->ipq_nfrags;
+ ip_freef(ipq[i].next);
+ }
+ }
+ }
ipflow_slowtimo();
- splx(s);
+ lck_mtx_unlock(ip_mutex);
}
/*
{
int i;
+ lck_mtx_lock(ip_mutex);
for (i = 0; i < IPREASS_NHASH; i++) {
while (ipq[i].next != &ipq[i]) {
- ipstat.ips_fragdropped++;
+ ipstat.ips_fragdropped += ipq[i].next->ipq_nfrags;
ip_freef(ipq[i].next);
}
}
+ lck_mtx_unlock(ip_mutex);
in_rtqdrain();
}
* Do option processing on a datagram,
* possibly discarding it if bad options are encountered,
* or forwarding it if source-routed.
+ * The pass argument is used when operating in the IPSTEALTH
+ * mode to tell what options to process:
+ * [LS]SRR (pass 0) or the others (pass 1).
+ * The reason for as many as two passes is that when doing IPSTEALTH,
+ * non-routing options should be processed only if the packet is for us.
* Returns 1 if packet has been forwarded/freed,
* 0 if the packet should be processed further.
*/
static int
-ip_dooptions(m)
- struct mbuf *m;
+ip_dooptions(struct mbuf *m, int pass, struct sockaddr_in *next_hop, struct route *ipforward_rt)
{
register struct ip *ip = mtod(m, struct ip *);
register u_char *cp;
optlen = 1;
else {
if (cnt < IPOPT_OLEN + sizeof(*cp)) {
- code = &cp[IPOPT_OLEN] - (u_char *)ip;
+ code = &cp[IPOPT_OLEN] - (u_char *)ip;
goto bad;
}
optlen = cp[IPOPT_OLEN];
*/
case IPOPT_LSRR:
case IPOPT_SSRR:
+ if (optlen < IPOPT_OFFSET + sizeof(*cp)) {
+ code = &cp[IPOPT_OLEN] - (u_char *)ip;
+ goto bad;
+ }
if ((off = cp[IPOPT_OFFSET]) < IPOPT_MINOFF) {
code = &cp[IPOPT_OFFSET] - (u_char *)ip;
goto bad;
*/
break;
}
+ else {
+ ifafree(&ia->ia_ifa);
+ ia = NULL;
+ }
off--; /* 0 origin */
- if (off > optlen - sizeof(struct in_addr)) {
+ if (off > optlen - (int)sizeof(struct in_addr)) {
/*
* End of source route. Should be for us.
*/
if (!ip_dosourceroute) {
if (ipforwarding) {
- char buf[16]; /* aaa.bbb.ccc.ddd\0 */
+ char buf[MAX_IPv4_STR_LEN];
+ char buf2[MAX_IPv4_STR_LEN];
/*
* Acting as a router, so generate ICMP
*/
nosourcerouting:
- strcpy(buf, inet_ntoa(ip->ip_dst));
- log(LOG_WARNING,
+ log(LOG_WARNING,
"attempted source route from %s to %s\n",
- inet_ntoa(ip->ip_src), buf);
+ inet_ntop(AF_INET, &ip->ip_src, buf, sizeof(buf)),
+ inet_ntop(AF_INET, &ip->ip_dst, buf2, sizeof(buf2)));
type = ICMP_UNREACH;
code = ICMP_UNREACH_SRCFAIL;
goto bad;
if (opt == IPOPT_SSRR) {
#define INA struct in_ifaddr *
#define SA struct sockaddr *
- if ((ia = (INA)ifa_ifwithdstaddr((SA)&ipaddr)) == 0)
- ia = (INA)ifa_ifwithnet((SA)&ipaddr);
- } else
- ia = ip_rtaddr(ipaddr.sin_addr);
+ if ((ia = (INA)ifa_ifwithdstaddr((SA)&ipaddr)) == 0) {
+ ia = (INA)ifa_ifwithnet((SA)&ipaddr);
+ }
+ } else {
+ ia = ip_rtaddr(ipaddr.sin_addr, ipforward_rt);
+ }
if (ia == 0) {
type = ICMP_UNREACH;
code = ICMP_UNREACH_SRCFAIL;
ip->ip_dst = ipaddr.sin_addr;
(void)memcpy(cp + off, &(IA_SIN(ia)->sin_addr),
sizeof(struct in_addr));
+ ifafree(&ia->ia_ifa);
+ ia = NULL;
cp[IPOPT_OFFSET] += sizeof(struct in_addr);
/*
* Let ip_intr's mcast routing check handle mcast pkts
* If no space remains, ignore.
*/
off--; /* 0 origin */
- if (off > optlen - sizeof(struct in_addr))
+ if (off > optlen - (int)sizeof(struct in_addr))
break;
(void)memcpy(&ipaddr.sin_addr, &ip->ip_dst,
sizeof(ipaddr.sin_addr));
* locate outgoing interface; if we're the destination,
* use the incoming interface (should be same).
*/
- if ((ia = (INA)ifa_ifwithaddr((SA)&ipaddr)) == 0 &&
- (ia = ip_rtaddr(ipaddr.sin_addr)) == 0) {
- type = ICMP_UNREACH;
- code = ICMP_UNREACH_HOST;
- goto bad;
+ if ((ia = (INA)ifa_ifwithaddr((SA)&ipaddr)) == 0) {
+ if ((ia = ip_rtaddr(ipaddr.sin_addr, ipforward_rt)) == 0) {
+ type = ICMP_UNREACH;
+ code = ICMP_UNREACH_HOST;
+ goto bad;
+ }
}
(void)memcpy(cp + off, &(IA_SIN(ia)->sin_addr),
sizeof(struct in_addr));
+ ifafree(&ia->ia_ifa);
+ ia = NULL;
cp[IPOPT_OFFSET] += sizeof(struct in_addr);
break;
case IPOPT_TS:
code = cp - (u_char *)ip;
ipt = (struct ip_timestamp *)cp;
- if (ipt->ipt_len < 5)
+ if (ipt->ipt_len < 4 || ipt->ipt_len > 40) {
+ code = (u_char *)&ipt->ipt_len - (u_char *)ip;
+ goto bad;
+ }
+ if (ipt->ipt_ptr < 5) {
+ code = (u_char *)&ipt->ipt_ptr - (u_char *)ip;
goto bad;
- if (ipt->ipt_ptr > ipt->ipt_len - sizeof(int32_t)) {
- if (++ipt->ipt_oflw == 0)
+ }
+ if (ipt->ipt_ptr >
+ ipt->ipt_len - (int)sizeof(int32_t)) {
+ if (++ipt->ipt_oflw == 0) {
+ code = (u_char *)&ipt->ipt_ptr -
+ (u_char *)ip;
goto bad;
+ }
break;
}
sin = (struct in_addr *)(cp + ipt->ipt_ptr - 1);
case IPOPT_TS_TSANDADDR:
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;
+ }
ipaddr.sin_addr = dst;
ia = (INA)ifaof_ifpforaddr((SA)&ipaddr,
m->m_pkthdr.rcvif);
(void)memcpy(sin, &IA_SIN(ia)->sin_addr,
sizeof(struct in_addr));
ipt->ipt_ptr += sizeof(struct in_addr);
+ ifafree(&ia->ia_ifa);
+ ia = NULL;
break;
case IPOPT_TS_PRESPEC:
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;
+ }
(void)memcpy(&ipaddr.sin_addr, sin,
sizeof(struct in_addr));
- if (ifa_ifwithaddr((SA)&ipaddr) == 0)
+ if ((ia = (struct in_ifaddr*)ifa_ifwithaddr((SA)&ipaddr)) == 0)
continue;
+ ifafree(&ia->ia_ifa);
+ ia = NULL;
ipt->ipt_ptr += sizeof(struct in_addr);
break;
default:
+ /* XXX can't take &ipt->ipt_flg */
+ code = (u_char *)&ipt->ipt_ptr -
+ (u_char *)ip + 1;
goto bad;
}
ntime = iptime();
}
}
if (forward && ipforwarding) {
- ip_forward(m, 1);
+ ip_forward(m, 1, next_hop, ipforward_rt);
return (1);
}
return (0);
bad:
ip->ip_len -= IP_VHL_HL(ip->ip_vhl) << 2; /* XXX icmp_error adds in hdr length */
+ lck_mtx_unlock(ip_mutex);
icmp_error(m, type, code, 0, 0);
+ lck_mtx_lock(ip_mutex);
ipstat.ips_badoptions++;
return (1);
}
* Given address of next destination (final or next hop),
* return internet address info of interface to be used to get there.
*/
-static struct in_ifaddr *
-ip_rtaddr(dst)
- struct in_addr dst;
+struct in_ifaddr *
+ip_rtaddr(dst, rt)
+ struct in_addr dst;
+ struct route *rt;
{
register struct sockaddr_in *sin;
- sin = (struct sockaddr_in *) &ipforward_rt.ro_dst;
+ sin = (struct sockaddr_in *)&rt->ro_dst;
- if (ipforward_rt.ro_rt == 0 || dst.s_addr != sin->sin_addr.s_addr) {
- if (ipforward_rt.ro_rt) {
- RTFREE(ipforward_rt.ro_rt);
- ipforward_rt.ro_rt = 0;
+ lck_mtx_lock(rt_mtx);
+ if (rt->ro_rt == 0 || dst.s_addr != sin->sin_addr.s_addr ||
+ rt->ro_rt->generation_id != route_generation) {
+ if (rt->ro_rt) {
+ rtfree_locked(rt->ro_rt);
+ rt->ro_rt = 0;
}
sin->sin_family = AF_INET;
sin->sin_len = sizeof(*sin);
sin->sin_addr = dst;
- rtalloc_ign(&ipforward_rt, RTF_PRCLONING);
+ rtalloc_ign_locked(rt, RTF_PRCLONING);
}
- if (ipforward_rt.ro_rt == 0)
+ if (rt->ro_rt == 0) {
+ lck_mtx_unlock(rt_mtx);
return ((struct in_ifaddr *)0);
- return ((struct in_ifaddr *) ipforward_rt.ro_rt->rt_ifa);
+ }
+
+ if (rt->ro_rt->rt_ifa)
+ ifaref(rt->ro_rt->rt_ifa);
+ lck_mtx_unlock(rt_mtx);
+ return ((struct in_ifaddr *) rt->ro_rt->rt_ifa);
}
/*
EHOSTUNREACH, EHOSTUNREACH, ECONNREFUSED, ECONNREFUSED,
EMSGSIZE, EHOSTUNREACH, 0, 0,
0, 0, 0, 0,
- ENOPROTOOPT
+ ENOPROTOOPT, ECONNREFUSED
};
/*
* The srcrt parameter indicates whether the packet is being forwarded
* via a source route.
*/
-#ifndef NATPT
-static
-#endif
-void
-ip_forward(m, srcrt)
- struct mbuf *m;
- int srcrt;
+static void
+ip_forward(struct mbuf *m, int srcrt, struct sockaddr_in *next_hop, struct route *ipforward_rt)
{
register struct ip *ip = mtod(m, struct ip *);
register struct sockaddr_in *sin;
int error, type = 0, code = 0;
struct mbuf *mcopy;
n_long dest;
+ struct in_addr pkt_dst;
struct ifnet *destifp;
#if IPSEC
struct ifnet dummyifp;
#endif
dest = 0;
+ /*
+ * Cache the destination address of the packet; this may be
+ * changed by use of 'ipfw fwd'.
+ */
+ pkt_dst = next_hop ? next_hop->sin_addr : ip->ip_dst;
+
#if DIAGNOSTIC
if (ipprintfs)
printf("forward: src %lx dst %lx ttl %x\n",
- (u_long)ip->ip_src.s_addr, (u_long)ip->ip_dst.s_addr,
+ (u_long)ip->ip_src.s_addr, (u_long)pkt_dst.s_addr,
ip->ip_ttl);
#endif
- if (m->m_flags & M_BCAST || in_canforward(ip->ip_dst) == 0) {
+ if (m->m_flags & (M_BCAST|M_MCAST) || in_canforward(pkt_dst) == 0) {
ipstat.ips_cantforward++;
m_freem(m);
return;
}
- HTONS(ip->ip_id);
- if (ip->ip_ttl <= IPTTLDEC) {
- icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS, dest, 0);
- return;
- }
- ip->ip_ttl -= IPTTLDEC;
-
-#if defined(PM)
- if (doRoute)
- {
- struct route *ipfw_rt;
-
- if ((ipfw_rt = pm_route(m)) != NULL)
- {
- mcopy = m_copy(m, 0, imin((int)ip->ip_len, 64));
-#if IPSEC
- ipsec_setsocket(m, NULL);
-#endif /*IPSEC*/
- error = ip_output(m, (struct mbuf *)0, ipfw_rt,
- IP_FORWARDING | IP_PROTOCOLROUTE , 0);
- goto clearAway;
- }
-
+#if IPSTEALTH
+ if (!ipstealth) {
+#endif
+ if (ip->ip_ttl <= IPTTLDEC) {
+ icmp_error(m, ICMP_TIMXCEED, ICMP_TIMXCEED_INTRANS,
+ dest, 0);
+ return;
+ }
+#if IPSTEALTH
}
#endif
- sin = (struct sockaddr_in *)&ipforward_rt.ro_dst;
- if ((rt = ipforward_rt.ro_rt) == 0 ||
- ip->ip_dst.s_addr != sin->sin_addr.s_addr) {
- if (ipforward_rt.ro_rt) {
- RTFREE(ipforward_rt.ro_rt);
- ipforward_rt.ro_rt = 0;
+ sin = (struct sockaddr_in *)&ipforward_rt->ro_dst;
+ if ((rt = ipforward_rt->ro_rt) == 0 ||
+ pkt_dst.s_addr != sin->sin_addr.s_addr ||
+ ipforward_rt->ro_rt->generation_id != route_generation) {
+ if (ipforward_rt->ro_rt) {
+ 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;
+ sin->sin_addr = pkt_dst;
- rtalloc_ign(&ipforward_rt, RTF_PRCLONING);
- if (ipforward_rt.ro_rt == 0) {
+ rtalloc_ign(ipforward_rt, RTF_PRCLONING);
+ if (ipforward_rt->ro_rt == 0) {
icmp_error(m, ICMP_UNREACH, ICMP_UNREACH_HOST, dest, 0);
return;
}
- rt = ipforward_rt.ro_rt;
+ rt = ipforward_rt->ro_rt;
}
/*
- * Save at most 64 bytes of the packet in case
- * we need to generate an ICMP message to the src.
+ * Save the IP header and at most 8 bytes of the payload,
+ * in case we need to generate an ICMP message to the src.
+ *
+ * We don't use m_copy() because it might return a reference
+ * to a shared cluster. Both this function and ip_output()
+ * assume exclusive access to the IP header in `m', so any
+ * data in a cluster may change before we reach icmp_error().
*/
- mcopy = m_copy(m, 0, imin((int)ip->ip_len, 64));
+ MGET(mcopy, M_DONTWAIT, m->m_type);
+ if (mcopy != NULL) {
+ M_COPY_PKTHDR(mcopy, m);
+ mcopy->m_len = imin((IP_VHL_HL(ip->ip_vhl) << 2) + 8,
+ (int)ip->ip_len);
+ m_copydata(m, 0, mcopy->m_len, mtod(mcopy, caddr_t));
+ }
+
+#if IPSTEALTH
+ if (!ipstealth) {
+#endif
+ ip->ip_ttl -= IPTTLDEC;
+#if IPSTEALTH
+ }
+#endif
/*
* If forwarding packet using same interface that it came in on,
if (rt->rt_flags & RTF_GATEWAY)
dest = satosin(rt->rt_gateway)->sin_addr.s_addr;
else
- dest = ip->ip_dst.s_addr;
+ dest = pkt_dst.s_addr;
/* Router requirements says to only send host redirects */
type = ICMP_REDIRECT;
code = ICMP_REDIRECT_HOST;
}
}
-#if IPSEC
- ipsec_setsocket(m, NULL);
-#endif /*IPSEC*/
- error = ip_output(m, (struct mbuf *)0, &ipforward_rt,
+ {
+ if (next_hop) {
+ /* Pass IPFORWARD info if available */
+ struct m_tag *tag;
+ struct ip_fwd_tag *ipfwd_tag;
+
+ tag = m_tag_alloc(KERNEL_MODULE_TAG_ID, KERNEL_TAG_TYPE_IPFORWARD,
+ sizeof(struct sockaddr_in), M_NOWAIT);
+ if (tag == NULL) {
+ error = ENOBUFS;
+ m_freem(m);
+ return;
+ }
+
+ ipfwd_tag = (struct ip_fwd_tag *)(tag+1);
+ ipfwd_tag->next_hop = next_hop;
+
+ m_tag_prepend(m, tag);
+ }
+ error = ip_output_list(m, 0, (struct mbuf *)0, ipforward_rt,
IP_FORWARDING, 0);
-#if defined(PM)
- clearAway:;
-#endif
+ }
if (error)
ipstat.ips_cantforward++;
else {
ipstat.ips_redirectsent++;
else {
if (mcopy) {
- ipflow_create(&ipforward_rt, mcopy);
+ ipflow_create(ipforward_rt, mcopy);
m_freem(mcopy);
}
return;
type = ICMP_UNREACH;
code = ICMP_UNREACH_NEEDFRAG;
#ifndef IPSEC
- if (ipforward_rt.ro_rt)
- destifp = ipforward_rt.ro_rt->rt_ifp;
+ if (ipforward_rt->ro_rt)
+ destifp = ipforward_rt->ro_rt->rt_ifp;
#else
/*
* If the packet is routed over IPsec tunnel, tell the
* tunnel MTU = if MTU - sizeof(IP) - ESP/AH hdrsiz
* XXX quickhack!!!
*/
- if (ipforward_rt.ro_rt) {
+ if (ipforward_rt->ro_rt) {
struct secpolicy *sp = NULL;
int ipsecerror;
int ipsechdr;
struct route *ro;
+ if (ipsec_bypass) {
+ destifp = ipforward_rt->ro_rt->rt_ifp;
+ ipstat.ips_cantfrag++;
+ break;
+ }
+ lck_mtx_lock(sadb_mutex);
sp = ipsec4_getpolicybyaddr(mcopy,
- IPSEC_DIR_OUTBOUND,
+ IPSEC_DIR_OUTBOUND,
IP_FORWARDING,
&ipsecerror);
if (sp == NULL)
- destifp = ipforward_rt.ro_rt->rt_ifp;
+ destifp = ipforward_rt->ro_rt->rt_ifp;
else {
/* count IPsec header size */
ipsechdr = ipsec4_hdrsiz(mcopy,
key_freesp(sp);
}
+ lck_mtx_unlock(sadb_mutex);
}
#endif /*IPSEC*/
ipstat.ips_cantfrag++;
type = ICMP_SOURCEQUENCH;
code = 0;
break;
+
+ case EACCES: /* ipfw denied packet */
+ m_freem(mcopy);
+ return;
}
icmp_error(mcopy, type, code, dest, destifp);
}
void
-ip_savecontrol(inp, mp, ip, m)
- register struct inpcb *inp;
- register struct mbuf **mp;
- register struct ip *ip;
- register struct mbuf *m;
+ip_savecontrol(
+ register struct inpcb *inp,
+ register struct mbuf **mp,
+ register struct ip *ip,
+ register struct mbuf *m)
{
if (inp->inp_socket->so_options & SO_TIMESTAMP) {
struct timeval tv;
struct sockaddr_dl *sdp;
struct sockaddr_dl *sdl2 = &sdlbuf.sdl;
+ ifnet_head_lock_shared();
if (((ifp = m->m_pkthdr.rcvif))
&& ( ifp->if_index && (ifp->if_index <= if_index))) {
- sdp = (struct sockaddr_dl *)(ifnet_addrs
- [ifp->if_index - 1]->ifa_addr);
+ struct ifaddr *ifa = ifnet_addrs[ifp->if_index - 1];
+
+ if (!ifa || !ifa->ifa_addr)
+ goto makedummy;
+
+ sdp = (struct sockaddr_dl *)ifa->ifa_addr;
/*
* Change our mind and don't try copy.
*/
sdl2->sdl_index = 0;
sdl2->sdl_nlen = sdl2->sdl_alen = sdl2->sdl_slen = 0;
}
+ ifnet_head_done();
*mp = sbcreatecontrol((caddr_t) sdl2, sdl2->sdl_len,
IP_RECVIF, IPPROTO_IP);
if (*mp)
mp = &(*mp)->m_next;
}
+ if (inp->inp_flags & INP_RECVTTL) {
+ *mp = sbcreatecontrol((caddr_t)&ip->ip_ttl, sizeof(ip->ip_ttl), IP_RECVTTL, IPPROTO_IP);
+ if (*mp) mp = &(*mp)->m_next;
+ }
}
int
projects / apple / xnu.git / blobdiff