freebsd-skq/sys/netinet6/ip6_forward.c
bz 5d8f0a53a7 Remove the single global unlocked route cache ip6_forward_rt
from the inet6 stack along with statistics and make sure we
properly free the rt in all cases.

While the current situation is not better performance wise it
prevents panics seen more often these days.
After more inet6 and ipsec cleanup we should be able to improve
the situation again passing the rt to ip6_forward directly.

Leave the ip6_forward_rt entry in struct vinet6 but mark it
for removal.

PR:		kern/128247, kern/131038
MFC after:	25 days
Committed from:	Bugathon #6
Tested by:	Denis Ahrens <denis@h3q.com> (different initial version)
2009-02-01 21:11:08 +00:00

629 lines
16 KiB
C

/*-
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the project nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $KAME: ip6_forward.c,v 1.69 2001/05/17 03:48:30 itojun Exp $
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_inet.h"
#include "opt_inet6.h"
#include "opt_ipsec.h"
#include "opt_ipstealth.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/domain.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/errno.h>
#include <sys/time.h>
#include <sys/kernel.h>
#include <sys/syslog.h>
#include <sys/vimage.h>
#include <net/if.h>
#include <net/route.h>
#include <net/pfil.h>
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#include <netinet6/in6_var.h>
#include <netinet/ip6.h>
#include <netinet6/ip6_var.h>
#include <netinet6/scope6_var.h>
#include <netinet/icmp6.h>
#include <netinet6/nd6.h>
#include <netinet6/vinet6.h>
#include <netinet/in_pcb.h>
#ifdef IPSEC
#include <netipsec/ipsec.h>
#include <netipsec/ipsec6.h>
#include <netipsec/key.h>
#endif /* IPSEC */
#include <netinet6/ip6protosw.h>
/*
* Forward a packet. If some error occurs return the sender
* an icmp packet. Note we can't always generate a meaningful
* icmp message because icmp doesn't have a large enough repertoire
* of codes and types.
*
* If not forwarding, just drop the packet. This could be confusing
* if ipforwarding was zero but some routing protocol was advancing
* us as a gateway to somewhere. However, we must let the routing
* protocol deal with that.
*
*/
void
ip6_forward(struct mbuf *m, int srcrt)
{
INIT_VNET_INET6(curvnet);
struct ip6_hdr *ip6 = mtod(m, struct ip6_hdr *);
struct sockaddr_in6 *dst = NULL;
struct rtentry *rt = NULL;
struct route_in6 rin6;
int error, type = 0, code = 0;
struct mbuf *mcopy = NULL;
struct ifnet *origifp; /* maybe unnecessary */
u_int32_t inzone, outzone;
struct in6_addr src_in6, dst_in6;
#ifdef IPSEC
INIT_VNET_IPSEC(curvnet);
struct secpolicy *sp = NULL;
int ipsecrt = 0;
#endif
char ip6bufs[INET6_ADDRSTRLEN], ip6bufd[INET6_ADDRSTRLEN];
#ifdef IPSEC
/*
* Check AH/ESP integrity.
*/
/*
* Don't increment ip6s_cantforward because this is the check
* before forwarding packet actually.
*/
if (ipsec6_in_reject(m, NULL)) {
V_ipsec6stat.in_polvio++;
m_freem(m);
return;
}
#endif /* IPSEC */
/*
* Do not forward packets to multicast destination (should be handled
* by ip6_mforward().
* Do not forward packets with unspecified source. It was discussed
* in July 2000, on the ipngwg mailing list.
*/
if ((m->m_flags & (M_BCAST|M_MCAST)) != 0 ||
IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst) ||
IN6_IS_ADDR_UNSPECIFIED(&ip6->ip6_src)) {
V_ip6stat.ip6s_cantforward++;
/* XXX in6_ifstat_inc(rt->rt_ifp, ifs6_in_discard) */
if (V_ip6_log_time + V_ip6_log_interval < time_second) {
V_ip6_log_time = time_second;
log(LOG_DEBUG,
"cannot forward "
"from %s to %s nxt %d received on %s\n",
ip6_sprintf(ip6bufs, &ip6->ip6_src),
ip6_sprintf(ip6bufd, &ip6->ip6_dst),
ip6->ip6_nxt,
if_name(m->m_pkthdr.rcvif));
}
m_freem(m);
return;
}
#ifdef IPSTEALTH
if (!V_ip6stealth) {
#endif
if (ip6->ip6_hlim <= IPV6_HLIMDEC) {
/* XXX in6_ifstat_inc(rt->rt_ifp, ifs6_in_discard) */
icmp6_error(m, ICMP6_TIME_EXCEEDED,
ICMP6_TIME_EXCEED_TRANSIT, 0);
return;
}
ip6->ip6_hlim -= IPV6_HLIMDEC;
#ifdef IPSTEALTH
}
#endif
/*
* Save at most ICMPV6_PLD_MAXLEN (= the min IPv6 MTU -
* size of IPv6 + ICMPv6 headers) bytes of the packet in case
* we need to generate an ICMP6 message to the src.
* Thanks to M_EXT, in most cases copy will not occur.
*
* It is important to save it before IPsec processing as IPsec
* processing may modify the mbuf.
*/
mcopy = m_copy(m, 0, imin(m->m_pkthdr.len, ICMPV6_PLD_MAXLEN));
#ifdef IPSEC
/* get a security policy for this packet */
sp = ipsec_getpolicybyaddr(m, IPSEC_DIR_OUTBOUND,
IP_FORWARDING, &error);
if (sp == NULL) {
V_ipsec6stat.out_inval++;
V_ip6stat.ip6s_cantforward++;
if (mcopy) {
#if 0
/* XXX: what icmp ? */
#else
m_freem(mcopy);
#endif
}
m_freem(m);
return;
}
error = 0;
/* check policy */
switch (sp->policy) {
case IPSEC_POLICY_DISCARD:
/*
* This packet is just discarded.
*/
V_ipsec6stat.out_polvio++;
V_ip6stat.ip6s_cantforward++;
KEY_FREESP(&sp);
if (mcopy) {
#if 0
/* XXX: what icmp ? */
#else
m_freem(mcopy);
#endif
}
m_freem(m);
return;
case IPSEC_POLICY_BYPASS:
case IPSEC_POLICY_NONE:
/* no need to do IPsec. */
KEY_FREESP(&sp);
goto skip_ipsec;
case IPSEC_POLICY_IPSEC:
if (sp->req == NULL) {
/* XXX should be panic ? */
printf("ip6_forward: No IPsec request specified.\n");
V_ip6stat.ip6s_cantforward++;
KEY_FREESP(&sp);
if (mcopy) {
#if 0
/* XXX: what icmp ? */
#else
m_freem(mcopy);
#endif
}
m_freem(m);
return;
}
/* do IPsec */
break;
case IPSEC_POLICY_ENTRUST:
default:
/* should be panic ?? */
printf("ip6_forward: Invalid policy found. %d\n", sp->policy);
KEY_FREESP(&sp);
goto skip_ipsec;
}
{
struct ipsecrequest *isr = NULL;
struct ipsec_output_state state;
/*
* when the kernel forwards a packet, it is not proper to apply
* IPsec transport mode to the packet is not proper. this check
* avoid from this.
* at present, if there is even a transport mode SA request in the
* security policy, the kernel does not apply IPsec to the packet.
* this check is not enough because the following case is valid.
* ipsec esp/tunnel/xxx-xxx/require esp/transport//require;
*/
for (isr = sp->req; isr; isr = isr->next) {
if (isr->saidx.mode == IPSEC_MODE_ANY)
goto doipsectunnel;
if (isr->saidx.mode == IPSEC_MODE_TUNNEL)
goto doipsectunnel;
}
/*
* if there's no need for tunnel mode IPsec, skip.
*/
if (!isr)
goto skip_ipsec;
doipsectunnel:
/*
* All the extension headers will become inaccessible
* (since they can be encrypted).
* Don't panic, we need no more updates to extension headers
* on inner IPv6 packet (since they are now encapsulated).
*
* IPv6 [ESP|AH] IPv6 [extension headers] payload
*/
bzero(&state, sizeof(state));
state.m = m;
state.ro = NULL; /* update at ipsec6_output_tunnel() */
state.dst = NULL; /* update at ipsec6_output_tunnel() */
error = ipsec6_output_tunnel(&state, sp, 0);
m = state.m;
KEY_FREESP(&sp);
if (error) {
/* mbuf is already reclaimed in ipsec6_output_tunnel. */
switch (error) {
case EHOSTUNREACH:
case ENETUNREACH:
case EMSGSIZE:
case ENOBUFS:
case ENOMEM:
break;
default:
printf("ip6_output (ipsec): error code %d\n", error);
/* FALLTHROUGH */
case ENOENT:
/* don't show these error codes to the user */
break;
}
V_ip6stat.ip6s_cantforward++;
if (mcopy) {
#if 0
/* XXX: what icmp ? */
#else
m_freem(mcopy);
#endif
}
m_freem(m);
return;
} else {
/*
* In the FAST IPSec case we have already
* re-injected the packet and it has been freed
* by the ipsec_done() function. So, just clean
* up after ourselves.
*/
m = NULL;
goto freecopy;
}
if ((m != NULL) && (ip6 != mtod(m, struct ip6_hdr *)) ){
/*
* now tunnel mode headers are added. we are originating
* packet instead of forwarding the packet.
*/
ip6_output(m, NULL, NULL, IPV6_FORWARDING/*XXX*/, NULL, NULL,
NULL);
goto freecopy;
}
/* adjust pointer */
dst = (struct sockaddr_in6 *)state.dst;
rt = state.ro ? state.ro->ro_rt : NULL;
if (dst != NULL && rt != NULL)
ipsecrt = 1;
}
if (ipsecrt)
goto skip_routing;
skip_ipsec:
#endif
bzero(&rin6, sizeof(struct route_in6));
dst = (struct sockaddr_in6 *)&rin6.ro_dst;
dst->sin6_len = sizeof(struct sockaddr_in6);
dst->sin6_family = AF_INET6;
dst->sin6_addr = ip6->ip6_dst;
rin6.ro_rt = rtalloc1((struct sockaddr *)dst, 0, 0);
if (rin6.ro_rt != NULL)
RT_UNLOCK(rin6.ro_rt);
else {
V_ip6stat.ip6s_noroute++;
in6_ifstat_inc(m->m_pkthdr.rcvif, ifs6_in_noroute);
if (mcopy) {
icmp6_error(mcopy, ICMP6_DST_UNREACH,
ICMP6_DST_UNREACH_NOROUTE, 0);
}
goto bad;
}
rt = rin6.ro_rt;
#ifdef IPSEC
skip_routing:
#endif
/*
* Source scope check: if a packet can't be delivered to its
* destination for the reason that the destination is beyond the scope
* of the source address, discard the packet and return an icmp6
* destination unreachable error with Code 2 (beyond scope of source
* address). We use a local copy of ip6_src, since in6_setscope()
* will possibly modify its first argument.
* [draft-ietf-ipngwg-icmp-v3-04.txt, Section 3.1]
*/
src_in6 = ip6->ip6_src;
if (in6_setscope(&src_in6, rt->rt_ifp, &outzone)) {
/* XXX: this should not happen */
V_ip6stat.ip6s_cantforward++;
V_ip6stat.ip6s_badscope++;
goto bad;
}
if (in6_setscope(&src_in6, m->m_pkthdr.rcvif, &inzone)) {
V_ip6stat.ip6s_cantforward++;
V_ip6stat.ip6s_badscope++;
goto bad;
}
if (inzone != outzone
#ifdef IPSEC
&& !ipsecrt
#endif
) {
V_ip6stat.ip6s_cantforward++;
V_ip6stat.ip6s_badscope++;
in6_ifstat_inc(rt->rt_ifp, ifs6_in_discard);
if (V_ip6_log_time + V_ip6_log_interval < time_second) {
V_ip6_log_time = time_second;
log(LOG_DEBUG,
"cannot forward "
"src %s, dst %s, nxt %d, rcvif %s, outif %s\n",
ip6_sprintf(ip6bufs, &ip6->ip6_src),
ip6_sprintf(ip6bufd, &ip6->ip6_dst),
ip6->ip6_nxt,
if_name(m->m_pkthdr.rcvif), if_name(rt->rt_ifp));
}
if (mcopy)
icmp6_error(mcopy, ICMP6_DST_UNREACH,
ICMP6_DST_UNREACH_BEYONDSCOPE, 0);
goto bad;
}
/*
* Destination scope check: if a packet is going to break the scope
* zone of packet's destination address, discard it. This case should
* usually be prevented by appropriately-configured routing table, but
* we need an explicit check because we may mistakenly forward the
* packet to a different zone by (e.g.) a default route.
*/
dst_in6 = ip6->ip6_dst;
if (in6_setscope(&dst_in6, m->m_pkthdr.rcvif, &inzone) != 0 ||
in6_setscope(&dst_in6, rt->rt_ifp, &outzone) != 0 ||
inzone != outzone) {
V_ip6stat.ip6s_cantforward++;
V_ip6stat.ip6s_badscope++;
goto bad;
}
if (m->m_pkthdr.len > IN6_LINKMTU(rt->rt_ifp)) {
in6_ifstat_inc(rt->rt_ifp, ifs6_in_toobig);
if (mcopy) {
u_long mtu;
#ifdef IPSEC
struct secpolicy *sp;
int ipsecerror;
size_t ipsechdrsiz;
#endif /* IPSEC */
mtu = IN6_LINKMTU(rt->rt_ifp);
#ifdef IPSEC
/*
* When we do IPsec tunnel ingress, we need to play
* with the link value (decrement IPsec header size
* from mtu value). The code is much simpler than v4
* case, as we have the outgoing interface for
* encapsulated packet as "rt->rt_ifp".
*/
sp = ipsec_getpolicybyaddr(mcopy, IPSEC_DIR_OUTBOUND,
IP_FORWARDING, &ipsecerror);
if (sp) {
ipsechdrsiz = ipsec6_hdrsiz(mcopy,
IPSEC_DIR_OUTBOUND, NULL);
if (ipsechdrsiz < mtu)
mtu -= ipsechdrsiz;
}
/*
* if mtu becomes less than minimum MTU,
* tell minimum MTU (and I'll need to fragment it).
*/
if (mtu < IPV6_MMTU)
mtu = IPV6_MMTU;
#endif /* IPSEC */
icmp6_error(mcopy, ICMP6_PACKET_TOO_BIG, 0, mtu);
}
goto bad;
}
if (rt->rt_flags & RTF_GATEWAY)
dst = (struct sockaddr_in6 *)rt->rt_gateway;
/*
* If we are to forward the packet using the same interface
* as one we got the packet from, perhaps we should send a redirect
* to sender to shortcut a hop.
* Only send redirect if source is sending directly to us,
* and if packet was not source routed (or has any options).
* Also, don't send redirect if forwarding using a route
* modified by a redirect.
*/
if (V_ip6_sendredirects && rt->rt_ifp == m->m_pkthdr.rcvif && !srcrt &&
#ifdef IPSEC
!ipsecrt &&
#endif /* IPSEC */
(rt->rt_flags & (RTF_DYNAMIC|RTF_MODIFIED)) == 0) {
if ((rt->rt_ifp->if_flags & IFF_POINTOPOINT) != 0) {
/*
* If the incoming interface is equal to the outgoing
* one, and the link attached to the interface is
* point-to-point, then it will be highly probable
* that a routing loop occurs. Thus, we immediately
* drop the packet and send an ICMPv6 error message.
*
* type/code is based on suggestion by Rich Draves.
* not sure if it is the best pick.
*/
icmp6_error(mcopy, ICMP6_DST_UNREACH,
ICMP6_DST_UNREACH_ADDR, 0);
goto bad;
}
type = ND_REDIRECT;
}
/*
* Fake scoped addresses. Note that even link-local source or
* destinaion can appear, if the originating node just sends the
* packet to us (without address resolution for the destination).
* Since both icmp6_error and icmp6_redirect_output fill the embedded
* link identifiers, we can do this stuff after making a copy for
* returning an error.
*/
if ((rt->rt_ifp->if_flags & IFF_LOOPBACK) != 0) {
/*
* See corresponding comments in ip6_output.
* XXX: but is it possible that ip6_forward() sends a packet
* to a loopback interface? I don't think so, and thus
* I bark here. (jinmei@kame.net)
* XXX: it is common to route invalid packets to loopback.
* also, the codepath will be visited on use of ::1 in
* rthdr. (itojun)
*/
#if 1
if (0)
#else
if ((rt->rt_flags & (RTF_BLACKHOLE|RTF_REJECT)) == 0)
#endif
{
printf("ip6_forward: outgoing interface is loopback. "
"src %s, dst %s, nxt %d, rcvif %s, outif %s\n",
ip6_sprintf(ip6bufs, &ip6->ip6_src),
ip6_sprintf(ip6bufd, &ip6->ip6_dst),
ip6->ip6_nxt, if_name(m->m_pkthdr.rcvif),
if_name(rt->rt_ifp));
}
/* we can just use rcvif in forwarding. */
origifp = m->m_pkthdr.rcvif;
}
else
origifp = rt->rt_ifp;
/*
* clear embedded scope identifiers if necessary.
* in6_clearscope will touch the addresses only when necessary.
*/
in6_clearscope(&ip6->ip6_src);
in6_clearscope(&ip6->ip6_dst);
/* Jump over all PFIL processing if hooks are not active. */
if (!PFIL_HOOKED(&inet6_pfil_hook))
goto pass;
/* Run through list of hooks for output packets. */
error = pfil_run_hooks(&inet6_pfil_hook, &m, rt->rt_ifp, PFIL_OUT, NULL);
if (error != 0)
goto senderr;
if (m == NULL)
goto freecopy;
ip6 = mtod(m, struct ip6_hdr *);
pass:
error = nd6_output(rt->rt_ifp, origifp, m, dst, rt);
if (error) {
in6_ifstat_inc(rt->rt_ifp, ifs6_out_discard);
V_ip6stat.ip6s_cantforward++;
} else {
V_ip6stat.ip6s_forward++;
in6_ifstat_inc(rt->rt_ifp, ifs6_out_forward);
if (type)
V_ip6stat.ip6s_redirectsent++;
else {
if (mcopy)
goto freecopy;
}
}
senderr:
if (mcopy == NULL)
goto out;
switch (error) {
case 0:
if (type == ND_REDIRECT) {
icmp6_redirect_output(mcopy, rt);
goto out;
}
goto freecopy;
case EMSGSIZE:
/* xxx MTU is constant in PPP? */
goto freecopy;
case ENOBUFS:
/* Tell source to slow down like source quench in IP? */
goto freecopy;
case ENETUNREACH: /* shouldn't happen, checked above */
case EHOSTUNREACH:
case ENETDOWN:
case EHOSTDOWN:
default:
type = ICMP6_DST_UNREACH;
code = ICMP6_DST_UNREACH_ADDR;
break;
}
icmp6_error(mcopy, type, code, 0);
goto out;
freecopy:
m_freem(mcopy);
goto out;
bad:
m_freem(m);
out:
if (rt != NULL
#ifdef IPSEC
&& !ipsecrt
#endif
)
RTFREE(rt);
}