NetBSD/sys/netinet/ip_etherip.c

265 lines
8.0 KiB
C
Raw Normal View History

Eliminate address family-specific route caches (struct route, struct route_in6, struct route_iso), replacing all caches with a struct route. The principle benefit of this change is that all of the protocol families can benefit from route cache-invalidation, which is necessary for correct routing. Route-cache invalidation fixes an ancient PR, kern/3508, at long last; it fixes various other PRs, also. Discussions with and ideas from Joerg Sonnenberger influenced this work tremendously. Of course, all design oversights and bugs are mine. DETAILS 1 I added to each address family a pool of sockaddrs. I have introduced routines for allocating, copying, and duplicating, and freeing sockaddrs: struct sockaddr *sockaddr_alloc(sa_family_t af, int flags); struct sockaddr *sockaddr_copy(struct sockaddr *dst, const struct sockaddr *src); struct sockaddr *sockaddr_dup(const struct sockaddr *src, int flags); void sockaddr_free(struct sockaddr *sa); sockaddr_alloc() returns either a sockaddr from the pool belonging to the specified family, or NULL if the pool is exhausted. The returned sockaddr has the right size for that family; sa_family and sa_len fields are initialized to the family and sockaddr length---e.g., sa_family = AF_INET and sa_len = sizeof(struct sockaddr_in). sockaddr_free() puts the given sockaddr back into its family's pool. sockaddr_dup() and sockaddr_copy() work analogously to strdup() and strcpy(), respectively. sockaddr_copy() KASSERTs that the family of the destination and source sockaddrs are alike. The 'flags' argumet for sockaddr_alloc() and sockaddr_dup() is passed directly to pool_get(9). 2 I added routines for initializing sockaddrs in each address family, sockaddr_in_init(), sockaddr_in6_init(), sockaddr_iso_init(), etc. They are fairly self-explanatory. 3 structs route_in6 and route_iso are no more. All protocol families use struct route. I have changed the route cache, 'struct route', so that it does not contain storage space for a sockaddr. Instead, struct route points to a sockaddr coming from the pool the sockaddr belongs to. I added a new method to struct route, rtcache_setdst(), for setting the cache destination: int rtcache_setdst(struct route *, const struct sockaddr *); rtcache_setdst() returns 0 on success, or ENOMEM if no memory is available to create the sockaddr storage. It is now possible for rtcache_getdst() to return NULL if, say, rtcache_setdst() failed. I check the return value for NULL everywhere in the kernel. 4 Each routing domain (struct domain) has a list of live route caches, dom_rtcache. rtflushall(sa_family_t af) looks up the domain indicated by 'af', walks the domain's list of route caches and invalidates each one.
2007-05-03 00:40:22 +04:00
/* $NetBSD: ip_etherip.c,v 1.5 2007/05/02 20:40:24 dyoung Exp $ */
/*
* Copyright (c) 2006, Hans Rosenfeld <rosenfeld@grumpf.hope-2000.org>
* 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 Hans Rosenfeld nor the names of his
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR 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 AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
*
* Copyright (C) 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.
*/
#include <sys/cdefs.h>
#include "opt_inet.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/socket.h>
#include <sys/sockio.h>
#include <sys/mbuf.h>
#include <sys/device.h>
#include <sys/errno.h>
#include <sys/ioctl.h>
#include <sys/syslog.h>
#include <sys/protosw.h>
#include <sys/kernel.h>
#include <net/if.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/ip_var.h>
#include <netinet/in_var.h>
#include <netinet/ip_etherip.h>
#include <net/if_ether.h>
#include <net/if_media.h>
#include <net/if_etherip.h>
#include <machine/stdarg.h>
int
ip_etherip_output(struct ifnet *ifp, struct mbuf *m)
{
struct etherip_softc *sc = (struct etherip_softc*)ifp->if_softc;
struct sockaddr_in *sin_src, *sin_dst;
struct ip iphdr; /* capsule IP header, host byte ordered */
struct etherip_header eiphdr;
int proto, error;
Eliminate address family-specific route caches (struct route, struct route_in6, struct route_iso), replacing all caches with a struct route. The principle benefit of this change is that all of the protocol families can benefit from route cache-invalidation, which is necessary for correct routing. Route-cache invalidation fixes an ancient PR, kern/3508, at long last; it fixes various other PRs, also. Discussions with and ideas from Joerg Sonnenberger influenced this work tremendously. Of course, all design oversights and bugs are mine. DETAILS 1 I added to each address family a pool of sockaddrs. I have introduced routines for allocating, copying, and duplicating, and freeing sockaddrs: struct sockaddr *sockaddr_alloc(sa_family_t af, int flags); struct sockaddr *sockaddr_copy(struct sockaddr *dst, const struct sockaddr *src); struct sockaddr *sockaddr_dup(const struct sockaddr *src, int flags); void sockaddr_free(struct sockaddr *sa); sockaddr_alloc() returns either a sockaddr from the pool belonging to the specified family, or NULL if the pool is exhausted. The returned sockaddr has the right size for that family; sa_family and sa_len fields are initialized to the family and sockaddr length---e.g., sa_family = AF_INET and sa_len = sizeof(struct sockaddr_in). sockaddr_free() puts the given sockaddr back into its family's pool. sockaddr_dup() and sockaddr_copy() work analogously to strdup() and strcpy(), respectively. sockaddr_copy() KASSERTs that the family of the destination and source sockaddrs are alike. The 'flags' argumet for sockaddr_alloc() and sockaddr_dup() is passed directly to pool_get(9). 2 I added routines for initializing sockaddrs in each address family, sockaddr_in_init(), sockaddr_in6_init(), sockaddr_iso_init(), etc. They are fairly self-explanatory. 3 structs route_in6 and route_iso are no more. All protocol families use struct route. I have changed the route cache, 'struct route', so that it does not contain storage space for a sockaddr. Instead, struct route points to a sockaddr coming from the pool the sockaddr belongs to. I added a new method to struct route, rtcache_setdst(), for setting the cache destination: int rtcache_setdst(struct route *, const struct sockaddr *); rtcache_setdst() returns 0 on success, or ENOMEM if no memory is available to create the sockaddr storage. It is now possible for rtcache_getdst() to return NULL if, say, rtcache_setdst() failed. I check the return value for NULL everywhere in the kernel. 4 Each routing domain (struct domain) has a list of live route caches, dom_rtcache. rtflushall(sa_family_t af) looks up the domain indicated by 'af', walks the domain's list of route caches and invalidates each one.
2007-05-03 00:40:22 +04:00
union {
struct sockaddr dst;
struct sockaddr_in dst4;
} u;
sin_src = (struct sockaddr_in *)sc->sc_src;
sin_dst = (struct sockaddr_in *)sc->sc_dst;
if (sin_src == NULL ||
sin_dst == NULL ||
sin_src->sin_family != AF_INET ||
sin_dst->sin_family != AF_INET) {
m_freem(m);
return EAFNOSUPPORT;
}
/* reset broadcast/multicast flags */
m->m_flags &= ~(M_BCAST|M_MCAST);
m->m_flags |= M_PKTHDR;
proto = IPPROTO_ETHERIP;
/* fill and prepend Ethernet-in-IP header */
eiphdr.eip_ver = ETHERIP_VERSION & ETHERIP_VER_VERS_MASK;
eiphdr.eip_pad = 0;
M_PREPEND(m, sizeof(struct etherip_header), M_DONTWAIT);
if (m == NULL)
return ENOBUFS;
if (M_UNWRITABLE(m, sizeof(struct etherip_header))) {
m = m_pullup(m, sizeof(struct etherip_header));
if (m == NULL)
return ENOBUFS;
}
memcpy(mtod(m, struct etherip_header *), &eiphdr,
sizeof(struct etherip_header));
/* fill new IP header */
memset(&iphdr, 0, sizeof(struct ip));
iphdr.ip_src = sin_src->sin_addr;
/* bidirectional configured tunnel mode */
if (sin_dst->sin_addr.s_addr != INADDR_ANY)
iphdr.ip_dst = sin_dst->sin_addr;
else {
m_freem(m);
return ENETUNREACH;
}
iphdr.ip_p = proto;
/* version will be set in ip_output() */
iphdr.ip_ttl = ETHERIP_TTL;
iphdr.ip_len = htons(m->m_pkthdr.len + sizeof(struct ip));
/* prepend new IP header */
M_PREPEND(m, sizeof(struct ip), M_DONTWAIT);
if (m == NULL)
return ENOBUFS;
if (M_UNWRITABLE(m, sizeof(struct ip)))
m = m_pullup(m, sizeof(struct ip));
memcpy(mtod(m, struct ip *), &iphdr, sizeof(struct ip));
Eliminate address family-specific route caches (struct route, struct route_in6, struct route_iso), replacing all caches with a struct route. The principle benefit of this change is that all of the protocol families can benefit from route cache-invalidation, which is necessary for correct routing. Route-cache invalidation fixes an ancient PR, kern/3508, at long last; it fixes various other PRs, also. Discussions with and ideas from Joerg Sonnenberger influenced this work tremendously. Of course, all design oversights and bugs are mine. DETAILS 1 I added to each address family a pool of sockaddrs. I have introduced routines for allocating, copying, and duplicating, and freeing sockaddrs: struct sockaddr *sockaddr_alloc(sa_family_t af, int flags); struct sockaddr *sockaddr_copy(struct sockaddr *dst, const struct sockaddr *src); struct sockaddr *sockaddr_dup(const struct sockaddr *src, int flags); void sockaddr_free(struct sockaddr *sa); sockaddr_alloc() returns either a sockaddr from the pool belonging to the specified family, or NULL if the pool is exhausted. The returned sockaddr has the right size for that family; sa_family and sa_len fields are initialized to the family and sockaddr length---e.g., sa_family = AF_INET and sa_len = sizeof(struct sockaddr_in). sockaddr_free() puts the given sockaddr back into its family's pool. sockaddr_dup() and sockaddr_copy() work analogously to strdup() and strcpy(), respectively. sockaddr_copy() KASSERTs that the family of the destination and source sockaddrs are alike. The 'flags' argumet for sockaddr_alloc() and sockaddr_dup() is passed directly to pool_get(9). 2 I added routines for initializing sockaddrs in each address family, sockaddr_in_init(), sockaddr_in6_init(), sockaddr_iso_init(), etc. They are fairly self-explanatory. 3 structs route_in6 and route_iso are no more. All protocol families use struct route. I have changed the route cache, 'struct route', so that it does not contain storage space for a sockaddr. Instead, struct route points to a sockaddr coming from the pool the sockaddr belongs to. I added a new method to struct route, rtcache_setdst(), for setting the cache destination: int rtcache_setdst(struct route *, const struct sockaddr *); rtcache_setdst() returns 0 on success, or ENOMEM if no memory is available to create the sockaddr storage. It is now possible for rtcache_getdst() to return NULL if, say, rtcache_setdst() failed. I check the return value for NULL everywhere in the kernel. 4 Each routing domain (struct domain) has a list of live route caches, dom_rtcache. rtflushall(sa_family_t af) looks up the domain indicated by 'af', walks the domain's list of route caches and invalidates each one.
2007-05-03 00:40:22 +04:00
sockaddr_in_init(&u.dst4, &sin_dst->sin_addr, 0);
if (rtcache_lookup(&sc->sc_ro, &u.dst) == NULL) {
m_freem(m);
return ENETUNREACH;
}
/* if it constitutes infinite encapsulation, punt. */
if (sc->sc_ro.ro_rt->rt_ifp == ifp) {
rtcache_free(&sc->sc_ro);
m_freem(m);
return ENETUNREACH; /*XXX*/
}
error = ip_output(m, NULL, &sc->sc_ro, 0, NULL, NULL);
return error;
}
void
ip_etherip_input(struct mbuf *m, ...)
{
struct etherip_softc *sc;
const struct ip *ip;
struct sockaddr_in *src, *dst;
struct ifnet *ifp = NULL;
int off, proto;
va_list ap;
va_start(ap, m);
off = va_arg(ap, int);
proto = va_arg(ap, int);
va_end(ap);
if (proto != IPPROTO_ETHERIP) {
m_freem(m);
ipstat.ips_noproto++;
return;
}
ip = mtod(m, const struct ip *);
/* find device configured for this packets src and dst */
LIST_FOREACH(sc, &etherip_softc_list, etherip_list) {
if (!sc->sc_src || !sc->sc_dst)
continue;
if (sc->sc_src->sa_family != AF_INET ||
sc->sc_dst->sa_family != AF_INET)
continue;
src = (struct sockaddr_in *)sc->sc_src;
dst = (struct sockaddr_in *)sc->sc_dst;
if (src->sin_addr.s_addr != ip->ip_dst.s_addr ||
dst->sin_addr.s_addr != ip->ip_src.s_addr)
continue;
ifp = &sc->sc_ec.ec_if;
break;
}
/* no matching device found */
if (!ifp) {
m_freem(m);
ipstat.ips_odropped++;
return;
}
m_adj(m, off);
/*
* Section 4 of RFC 3378 requires that the EtherIP header of incoming
* packets is verified to contain the correct values in the version and
* reserved fields, and packets with wrong values be dropped.
*
* There is some discussion about what exactly the header should look
* like, the RFC is not very clear there. To be compatible with broken
* implementations, we don't check the header on incoming packets,
* relying on the ethernet code to filter out garbage.
*
* The header we use for sending is compatible with the original
* implementation in OpenBSD, which was used in former NetBSD versions
* and is used in FreeBSD. One Linux implementation is known to use the
* same value.
*/
m_adj(m, sizeof(struct etherip_header));
m = m_pullup(m, sizeof(struct ether_header));
if (m == NULL) {
ifp->if_ierrors++;
return;
}
m->m_pkthdr.rcvif = ifp;
m->m_flags &= ~(M_BCAST|M_MCAST);
#if NBPFILTER > 0
if (ifp->if_bpf)
bpf_mtap(ifp->if_bpf, m);
#endif
ifp->if_ipackets++;
(ifp->if_input)(ifp, m);
return;
}