diff --git a/sys/netinet/in.c b/sys/netinet/in.c
index 6bc4f638a98a..fccad3d9ede2 100644
--- a/sys/netinet/in.c
+++ b/sys/netinet/in.c
@@ -1,1699 +1,1717 @@
/*-
* SPDX-License-Identifier: BSD-3-Clause
*
* Copyright (c) 1982, 1986, 1991, 1993
* The Regents of the University of California. All rights reserved.
* Copyright (C) 2001 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 University 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 REGENTS 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 REGENTS 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.
*
* @(#)in.c 8.4 (Berkeley) 1/9/95
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include <sys/param.h>
#include <sys/eventhandler.h>
#include <sys/systm.h>
#include <sys/sockio.h>
#include <sys/malloc.h>
#include <sys/priv.h>
#include <sys/socket.h>
#include <sys/jail.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/proc.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/sx.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_arp.h>
#include <net/if_dl.h>
#include <net/if_llatbl.h>
#include <net/if_types.h>
#include <net/route.h>
#include <net/route/nhop.h>
#include <net/route/route_ctl.h>
#include <net/vnet.h>
#include <netinet/if_ether.h>
#include <netinet/in.h>
#include <netinet/in_fib.h>
#include <netinet/in_var.h>
#include <netinet/in_pcb.h>
#include <netinet/ip_var.h>
#include <netinet/ip_carp.h>
#include <netinet/igmp_var.h>
#include <netinet/udp.h>
#include <netinet/udp_var.h>
static int in_aifaddr_ioctl(u_long, caddr_t, struct ifnet *, struct thread *);
static int in_difaddr_ioctl(u_long, caddr_t, struct ifnet *, struct thread *);
static int in_gifaddr_ioctl(u_long, caddr_t, struct ifnet *, struct thread *);
static void in_socktrim(struct sockaddr_in *);
static void in_purgemaddrs(struct ifnet *);
static bool ia_need_loopback_route(const struct in_ifaddr *);
VNET_DEFINE_STATIC(int, nosameprefix);
#define V_nosameprefix VNET(nosameprefix)
SYSCTL_INT(_net_inet_ip, OID_AUTO, no_same_prefix, CTLFLAG_VNET | CTLFLAG_RW,
&VNET_NAME(nosameprefix), 0,
"Refuse to create same prefixes on different interfaces");
VNET_DEFINE_STATIC(bool, broadcast_lowest);
#define V_broadcast_lowest VNET(broadcast_lowest)
SYSCTL_BOOL(_net_inet_ip, OID_AUTO, broadcast_lowest, CTLFLAG_VNET | CTLFLAG_RW,
&VNET_NAME(broadcast_lowest), 0,
"Treat lowest address on a subnet (host 0) as broadcast");
VNET_DECLARE(struct inpcbinfo, ripcbinfo);
#define V_ripcbinfo VNET(ripcbinfo)
static struct sx in_control_sx;
SX_SYSINIT(in_control_sx, &in_control_sx, "in_control");
/*
* Return 1 if an internet address is for a ``local'' host
* (one to which we have a connection).
*/
int
in_localaddr(struct in_addr in)
{
u_long i = ntohl(in.s_addr);
struct in_ifaddr *ia;
NET_EPOCH_ASSERT();
CK_STAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
if ((i & ia->ia_subnetmask) == ia->ia_subnet)
return (1);
}
return (0);
}
/*
* Return 1 if an internet address is for the local host and configured
* on one of its interfaces.
*/
bool
in_localip(struct in_addr in)
{
struct in_ifaddr *ia;
NET_EPOCH_ASSERT();
CK_LIST_FOREACH(ia, INADDR_HASH(in.s_addr), ia_hash)
if (IA_SIN(ia)->sin_addr.s_addr == in.s_addr)
return (true);
return (false);
}
+/*
+ * Like in_localip(), but FIB-aware.
+ */
+bool
+in_localip_fib(struct in_addr in, uint16_t fib)
+{
+ struct in_ifaddr *ia;
+
+ NET_EPOCH_ASSERT();
+
+ CK_LIST_FOREACH(ia, INADDR_HASH(in.s_addr), ia_hash)
+ if (IA_SIN(ia)->sin_addr.s_addr == in.s_addr &&
+ ia->ia_ifa.ifa_ifp->if_fib == fib)
+ return (true);
+
+ return (false);
+}
+
/*
* Return 1 if an internet address is configured on an interface.
*/
int
in_ifhasaddr(struct ifnet *ifp, struct in_addr in)
{
struct ifaddr *ifa;
struct in_ifaddr *ia;
NET_EPOCH_ASSERT();
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET)
continue;
ia = (struct in_ifaddr *)ifa;
if (ia->ia_addr.sin_addr.s_addr == in.s_addr)
return (1);
}
return (0);
}
/*
* Return a reference to the interface address which is different to
* the supplied one but with same IP address value.
*/
static struct in_ifaddr *
in_localip_more(struct in_ifaddr *original_ia)
{
struct epoch_tracker et;
in_addr_t original_addr = IA_SIN(original_ia)->sin_addr.s_addr;
uint32_t original_fib = original_ia->ia_ifa.ifa_ifp->if_fib;
struct in_ifaddr *ia;
NET_EPOCH_ENTER(et);
CK_LIST_FOREACH(ia, INADDR_HASH(original_addr), ia_hash) {
in_addr_t addr = IA_SIN(ia)->sin_addr.s_addr;
uint32_t fib = ia->ia_ifa.ifa_ifp->if_fib;
if (!V_rt_add_addr_allfibs && (original_fib != fib))
continue;
if ((original_ia != ia) && (original_addr == addr)) {
ifa_ref(&ia->ia_ifa);
NET_EPOCH_EXIT(et);
return (ia);
}
}
NET_EPOCH_EXIT(et);
return (NULL);
}
/*
* Tries to find first IPv4 address in the provided fib.
* Prefers non-loopback addresses and return loopback IFF
* @loopback_ok is set.
*
* Returns ifa or NULL.
*/
struct in_ifaddr *
in_findlocal(uint32_t fibnum, bool loopback_ok)
{
struct in_ifaddr *ia = NULL, *ia_lo = NULL;
NET_EPOCH_ASSERT();
CK_STAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
uint32_t ia_fib = ia->ia_ifa.ifa_ifp->if_fib;
if (!V_rt_add_addr_allfibs && (fibnum != ia_fib))
continue;
if (!IN_LOOPBACK(ntohl(IA_SIN(ia)->sin_addr.s_addr)))
break;
if (loopback_ok)
ia_lo = ia;
}
if (ia == NULL)
ia = ia_lo;
return (ia);
}
/*
* Determine whether an IP address is in a reserved set of addresses
* that may not be forwarded, or whether datagrams to that destination
* may be forwarded.
*/
int
in_canforward(struct in_addr in)
{
u_long i = ntohl(in.s_addr);
if (IN_EXPERIMENTAL(i) || IN_MULTICAST(i) || IN_LINKLOCAL(i) ||
IN_ZERONET(i) || IN_LOOPBACK(i))
return (0);
return (1);
}
/*
* Trim a mask in a sockaddr
*/
static void
in_socktrim(struct sockaddr_in *ap)
{
char *cplim = (char *) &ap->sin_addr;
char *cp = (char *) (&ap->sin_addr + 1);
ap->sin_len = 0;
while (--cp >= cplim)
if (*cp) {
(ap)->sin_len = cp - (char *) (ap) + 1;
break;
}
}
/*
* Generic internet control operations (ioctl's).
*/
int
in_control(struct socket *so, u_long cmd, caddr_t data, struct ifnet *ifp,
struct thread *td)
{
struct ifreq *ifr = (struct ifreq *)data;
struct sockaddr_in *addr = (struct sockaddr_in *)&ifr->ifr_addr;
struct epoch_tracker et;
struct ifaddr *ifa;
struct in_ifaddr *ia;
int error;
if (ifp == NULL)
return (EADDRNOTAVAIL);
/*
* Filter out 4 ioctls we implement directly. Forward the rest
* to specific functions and ifp->if_ioctl().
*/
switch (cmd) {
case SIOCGIFADDR:
case SIOCGIFBRDADDR:
case SIOCGIFDSTADDR:
case SIOCGIFNETMASK:
break;
case SIOCGIFALIAS:
sx_xlock(&in_control_sx);
error = in_gifaddr_ioctl(cmd, data, ifp, td);
sx_xunlock(&in_control_sx);
return (error);
case SIOCDIFADDR:
sx_xlock(&in_control_sx);
error = in_difaddr_ioctl(cmd, data, ifp, td);
sx_xunlock(&in_control_sx);
return (error);
case OSIOCAIFADDR: /* 9.x compat */
case SIOCAIFADDR:
sx_xlock(&in_control_sx);
error = in_aifaddr_ioctl(cmd, data, ifp, td);
sx_xunlock(&in_control_sx);
return (error);
case SIOCSIFADDR:
case SIOCSIFBRDADDR:
case SIOCSIFDSTADDR:
case SIOCSIFNETMASK:
/* We no longer support that old commands. */
return (EINVAL);
default:
if (ifp->if_ioctl == NULL)
return (EOPNOTSUPP);
return ((*ifp->if_ioctl)(ifp, cmd, data));
}
if (addr->sin_addr.s_addr != INADDR_ANY &&
prison_check_ip4(td->td_ucred, &addr->sin_addr) != 0)
return (EADDRNOTAVAIL);
/*
* Find address for this interface, if it exists. If an
* address was specified, find that one instead of the
* first one on the interface, if possible.
*/
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET)
continue;
ia = (struct in_ifaddr *)ifa;
if (ia->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr)
break;
}
if (ifa == NULL)
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
if (ifa->ifa_addr->sa_family == AF_INET) {
ia = (struct in_ifaddr *)ifa;
if (prison_check_ip4(td->td_ucred,
&ia->ia_addr.sin_addr) == 0)
break;
}
if (ifa == NULL) {
NET_EPOCH_EXIT(et);
return (EADDRNOTAVAIL);
}
error = 0;
switch (cmd) {
case SIOCGIFADDR:
*addr = ia->ia_addr;
break;
case SIOCGIFBRDADDR:
if ((ifp->if_flags & IFF_BROADCAST) == 0) {
error = EINVAL;
break;
}
*addr = ia->ia_broadaddr;
break;
case SIOCGIFDSTADDR:
if ((ifp->if_flags & IFF_POINTOPOINT) == 0) {
error = EINVAL;
break;
}
*addr = ia->ia_dstaddr;
break;
case SIOCGIFNETMASK:
*addr = ia->ia_sockmask;
break;
}
NET_EPOCH_EXIT(et);
return (error);
}
static int
in_aifaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct thread *td)
{
const struct in_aliasreq *ifra = (struct in_aliasreq *)data;
const struct sockaddr_in *addr = &ifra->ifra_addr;
const struct sockaddr_in *broadaddr = &ifra->ifra_broadaddr;
const struct sockaddr_in *mask = &ifra->ifra_mask;
const struct sockaddr_in *dstaddr = &ifra->ifra_dstaddr;
const int vhid = (cmd == SIOCAIFADDR) ? ifra->ifra_vhid : 0;
struct epoch_tracker et;
struct ifaddr *ifa;
struct in_ifaddr *ia;
bool iaIsFirst;
int error = 0;
error = priv_check(td, PRIV_NET_ADDIFADDR);
if (error)
return (error);
/*
* ifra_addr must be present and be of INET family.
* ifra_broadaddr/ifra_dstaddr and ifra_mask are optional.
*/
if (addr->sin_len != sizeof(struct sockaddr_in) ||
addr->sin_family != AF_INET)
return (EINVAL);
if (broadaddr->sin_len != 0 &&
(broadaddr->sin_len != sizeof(struct sockaddr_in) ||
broadaddr->sin_family != AF_INET))
return (EINVAL);
if (mask->sin_len != 0 &&
(mask->sin_len != sizeof(struct sockaddr_in) ||
mask->sin_family != AF_INET))
return (EINVAL);
if ((ifp->if_flags & IFF_POINTOPOINT) &&
(dstaddr->sin_len != sizeof(struct sockaddr_in) ||
dstaddr->sin_addr.s_addr == INADDR_ANY))
return (EDESTADDRREQ);
if (vhid != 0 && carp_attach_p == NULL)
return (EPROTONOSUPPORT);
/*
* See whether address already exist.
*/
iaIsFirst = true;
ia = NULL;
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
struct in_ifaddr *it;
if (ifa->ifa_addr->sa_family != AF_INET)
continue;
it = (struct in_ifaddr *)ifa;
if (it->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr &&
prison_check_ip4(td->td_ucred, &addr->sin_addr) == 0)
ia = it;
else
iaIsFirst = false;
}
NET_EPOCH_EXIT(et);
if (ia != NULL)
(void )in_difaddr_ioctl(cmd, data, ifp, td);
ifa = ifa_alloc(sizeof(struct in_ifaddr), M_WAITOK);
ia = (struct in_ifaddr *)ifa;
ifa->ifa_addr = (struct sockaddr *)&ia->ia_addr;
ifa->ifa_dstaddr = (struct sockaddr *)&ia->ia_dstaddr;
ifa->ifa_netmask = (struct sockaddr *)&ia->ia_sockmask;
callout_init_rw(&ia->ia_garp_timer, &ifp->if_addr_lock,
CALLOUT_RETURNUNLOCKED);
ia->ia_ifp = ifp;
ia->ia_addr = *addr;
if (mask->sin_len != 0) {
ia->ia_sockmask = *mask;
ia->ia_subnetmask = ntohl(ia->ia_sockmask.sin_addr.s_addr);
} else {
/*
* If netmask isn't supplied, use default for now.
* This is deprecated for interfaces other than loopback
* or point-to-point; warn in other cases. In the future
* we should return an error rather than warning.
*/
if ((ifp->if_flags & (IFF_POINTOPOINT | IFF_LOOPBACK)) == 0)
printf("%s: set address: WARNING: network mask"
" should be specified; using default mask\n",
ifp->if_xname);
ia->ia_subnetmask = IN_NETMASK_DEFAULT;
ia->ia_sockmask.sin_addr.s_addr = htonl(ia->ia_subnetmask);
}
ia->ia_subnet = ntohl(addr->sin_addr.s_addr) & ia->ia_subnetmask;
in_socktrim(&ia->ia_sockmask);
if (ifp->if_flags & IFF_BROADCAST) {
if (broadaddr->sin_len != 0) {
ia->ia_broadaddr = *broadaddr;
} else if (ia->ia_subnetmask == IN_RFC3021_MASK) {
ia->ia_broadaddr.sin_addr.s_addr = INADDR_BROADCAST;
ia->ia_broadaddr.sin_len = sizeof(struct sockaddr_in);
ia->ia_broadaddr.sin_family = AF_INET;
} else {
ia->ia_broadaddr.sin_addr.s_addr =
htonl(ia->ia_subnet | ~ia->ia_subnetmask);
ia->ia_broadaddr.sin_len = sizeof(struct sockaddr_in);
ia->ia_broadaddr.sin_family = AF_INET;
}
}
if (ifp->if_flags & IFF_POINTOPOINT)
ia->ia_dstaddr = *dstaddr;
if (vhid != 0) {
error = (*carp_attach_p)(&ia->ia_ifa, vhid);
if (error)
return (error);
}
/* if_addrhead is already referenced by ifa_alloc() */
IF_ADDR_WLOCK(ifp);
CK_STAILQ_INSERT_TAIL(&ifp->if_addrhead, ifa, ifa_link);
IF_ADDR_WUNLOCK(ifp);
ifa_ref(ifa); /* in_ifaddrhead */
sx_assert(&in_control_sx, SA_XLOCKED);
CK_STAILQ_INSERT_TAIL(&V_in_ifaddrhead, ia, ia_link);
CK_LIST_INSERT_HEAD(INADDR_HASH(ia->ia_addr.sin_addr.s_addr), ia,
ia_hash);
/*
* Give the interface a chance to initialize
* if this is its first address,
* and to validate the address if necessary.
*/
if (ifp->if_ioctl != NULL) {
error = (*ifp->if_ioctl)(ifp, SIOCSIFADDR, (caddr_t)ia);
if (error)
goto fail1;
}
/*
* Add route for the network.
*/
if (vhid == 0) {
error = in_addprefix(ia);
if (error)
goto fail1;
}
/*
* Add a loopback route to self.
*/
if (vhid == 0 && ia_need_loopback_route(ia)) {
struct in_ifaddr *eia;
eia = in_localip_more(ia);
if (eia == NULL) {
error = ifa_add_loopback_route((struct ifaddr *)ia,
(struct sockaddr *)&ia->ia_addr);
if (error)
goto fail2;
} else
ifa_free(&eia->ia_ifa);
}
if (iaIsFirst && (ifp->if_flags & IFF_MULTICAST)) {
struct in_addr allhosts_addr;
struct in_ifinfo *ii;
ii = ((struct in_ifinfo *)ifp->if_afdata[AF_INET]);
allhosts_addr.s_addr = htonl(INADDR_ALLHOSTS_GROUP);
error = in_joingroup(ifp, &allhosts_addr, NULL,
&ii->ii_allhosts);
}
/*
* Note: we don't need extra reference for ifa, since we called
* with sx lock held, and ifaddr can not be deleted in concurrent
* thread.
*/
EVENTHANDLER_INVOKE(ifaddr_event_ext, ifp, ifa, IFADDR_EVENT_ADD);
return (error);
fail2:
if (vhid == 0)
(void )in_scrubprefix(ia, LLE_STATIC);
fail1:
if (ia->ia_ifa.ifa_carp)
(*carp_detach_p)(&ia->ia_ifa, false);
IF_ADDR_WLOCK(ifp);
CK_STAILQ_REMOVE(&ifp->if_addrhead, &ia->ia_ifa, ifaddr, ifa_link);
IF_ADDR_WUNLOCK(ifp);
ifa_free(&ia->ia_ifa); /* if_addrhead */
sx_assert(&in_control_sx, SA_XLOCKED);
CK_STAILQ_REMOVE(&V_in_ifaddrhead, ia, in_ifaddr, ia_link);
CK_LIST_REMOVE(ia, ia_hash);
ifa_free(&ia->ia_ifa); /* in_ifaddrhead */
return (error);
}
static int
in_difaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct thread *td)
{
const struct ifreq *ifr = (struct ifreq *)data;
const struct sockaddr_in *addr = (const struct sockaddr_in *)
&ifr->ifr_addr;
struct ifaddr *ifa;
struct in_ifaddr *ia;
bool deleteAny, iaIsLast;
int error;
if (td != NULL) {
error = priv_check(td, PRIV_NET_DELIFADDR);
if (error)
return (error);
}
if (addr->sin_len != sizeof(struct sockaddr_in) ||
addr->sin_family != AF_INET)
deleteAny = true;
else
deleteAny = false;
iaIsLast = true;
ia = NULL;
IF_ADDR_WLOCK(ifp);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
struct in_ifaddr *it;
if (ifa->ifa_addr->sa_family != AF_INET)
continue;
it = (struct in_ifaddr *)ifa;
if (deleteAny && ia == NULL && (td == NULL ||
prison_check_ip4(td->td_ucred, &it->ia_addr.sin_addr) == 0))
ia = it;
if (it->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr &&
(td == NULL || prison_check_ip4(td->td_ucred,
&addr->sin_addr) == 0))
ia = it;
if (it != ia)
iaIsLast = false;
}
if (ia == NULL) {
IF_ADDR_WUNLOCK(ifp);
return (EADDRNOTAVAIL);
}
CK_STAILQ_REMOVE(&ifp->if_addrhead, &ia->ia_ifa, ifaddr, ifa_link);
IF_ADDR_WUNLOCK(ifp);
ifa_free(&ia->ia_ifa); /* if_addrhead */
sx_assert(&in_control_sx, SA_XLOCKED);
CK_STAILQ_REMOVE(&V_in_ifaddrhead, ia, in_ifaddr, ia_link);
CK_LIST_REMOVE(ia, ia_hash);
/*
* in_scrubprefix() kills the interface route.
*/
in_scrubprefix(ia, LLE_STATIC);
/*
* in_ifadown gets rid of all the rest of
* the routes. This is not quite the right
* thing to do, but at least if we are running
* a routing process they will come back.
*/
in_ifadown(&ia->ia_ifa, 1);
if (ia->ia_ifa.ifa_carp)
(*carp_detach_p)(&ia->ia_ifa, cmd == SIOCAIFADDR);
/*
* If this is the last IPv4 address configured on this
* interface, leave the all-hosts group.
* No state-change report need be transmitted.
*/
if (iaIsLast && (ifp->if_flags & IFF_MULTICAST)) {
struct in_ifinfo *ii;
ii = ((struct in_ifinfo *)ifp->if_afdata[AF_INET]);
if (ii->ii_allhosts) {
(void)in_leavegroup(ii->ii_allhosts, NULL);
ii->ii_allhosts = NULL;
}
}
IF_ADDR_WLOCK(ifp);
if (callout_stop(&ia->ia_garp_timer) == 1) {
ifa_free(&ia->ia_ifa);
}
IF_ADDR_WUNLOCK(ifp);
EVENTHANDLER_INVOKE(ifaddr_event_ext, ifp, &ia->ia_ifa,
IFADDR_EVENT_DEL);
ifa_free(&ia->ia_ifa); /* in_ifaddrhead */
return (0);
}
static int
in_gifaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct thread *td)
{
struct in_aliasreq *ifra = (struct in_aliasreq *)data;
const struct sockaddr_in *addr = &ifra->ifra_addr;
struct epoch_tracker et;
struct ifaddr *ifa;
struct in_ifaddr *ia;
/*
* ifra_addr must be present and be of INET family.
*/
if (addr->sin_len != sizeof(struct sockaddr_in) ||
addr->sin_family != AF_INET)
return (EINVAL);
/*
* See whether address exist.
*/
ia = NULL;
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
struct in_ifaddr *it;
if (ifa->ifa_addr->sa_family != AF_INET)
continue;
it = (struct in_ifaddr *)ifa;
if (it->ia_addr.sin_addr.s_addr == addr->sin_addr.s_addr &&
prison_check_ip4(td->td_ucred, &addr->sin_addr) == 0) {
ia = it;
break;
}
}
if (ia == NULL) {
NET_EPOCH_EXIT(et);
return (EADDRNOTAVAIL);
}
ifra->ifra_mask = ia->ia_sockmask;
if ((ifp->if_flags & IFF_POINTOPOINT) &&
ia->ia_dstaddr.sin_family == AF_INET)
ifra->ifra_dstaddr = ia->ia_dstaddr;
else if ((ifp->if_flags & IFF_BROADCAST) &&
ia->ia_broadaddr.sin_family == AF_INET)
ifra->ifra_broadaddr = ia->ia_broadaddr;
else
memset(&ifra->ifra_broadaddr, 0,
sizeof(ifra->ifra_broadaddr));
NET_EPOCH_EXIT(et);
return (0);
}
static int
in_match_ifaddr(const struct rtentry *rt, const struct nhop_object *nh, void *arg)
{
if (nh->nh_ifa == (struct ifaddr *)arg)
return (1);
return (0);
}
static int
in_handle_prefix_route(uint32_t fibnum, int cmd,
struct sockaddr_in *dst, struct sockaddr_in *netmask, struct ifaddr *ifa,
struct ifnet *ifp)
{
NET_EPOCH_ASSERT();
/* Prepare gateway */
struct sockaddr_dl_short sdl = {
.sdl_family = AF_LINK,
.sdl_len = sizeof(struct sockaddr_dl_short),
.sdl_type = ifa->ifa_ifp->if_type,
.sdl_index = ifa->ifa_ifp->if_index,
};
struct rt_addrinfo info = {
.rti_ifa = ifa,
.rti_ifp = ifp,
.rti_flags = RTF_PINNED | ((netmask != NULL) ? 0 : RTF_HOST),
.rti_info = {
[RTAX_DST] = (struct sockaddr *)dst,
[RTAX_NETMASK] = (struct sockaddr *)netmask,
[RTAX_GATEWAY] = (struct sockaddr *)&sdl,
},
/* Ensure we delete the prefix IFF prefix ifa matches */
.rti_filter = in_match_ifaddr,
.rti_filterdata = ifa,
};
return (rib_handle_ifaddr_info(fibnum, cmd, &info));
}
/*
* Routing table interaction with interface addresses.
*
* In general, two types of routes needs to be installed:
* a) "interface" or "prefix" route, telling user that the addresses
* behind the ifa prefix are reached directly.
* b) "loopback" route installed for the ifa address, telling user that
* the address belongs to local system.
*
* Handling for (a) and (b) differs in multi-fib aspects, hence they
* are implemented in different functions below.
*
* The cases above may intersect - /32 interface aliases results in
* the same prefix produced by (a) and (b). This blurs the definition
* of the "loopback" route and complicate interactions. The interaction
* table is defined below. The case numbers are used in the multiple
* functions below to refer to the particular test case.
*
* There can be multiple options:
* 1) Adding address with prefix on non-p2p/non-loopback interface.
* Example: 192.0.2.1/24. Action:
* * add "prefix" route towards 192.0.2.0/24 via @ia interface,
* using @ia as an address source.
* * add "loopback" route towards 192.0.2.1 via V_loif, saving
* @ia ifp in the gateway and using @ia as an address source.
*
* 2) Adding address with /32 mask to non-p2p/non-loopback interface.
* Example: 192.0.2.2/32. Action:
* * add "prefix" host route via V_loif, using @ia as an address source.
*
* 3) Adding address with or without prefix to p2p interface.
* Example: 10.0.0.1/24->10.0.0.2. Action:
* * add "prefix" host route towards 10.0.0.2 via this interface, using @ia
* as an address source. Note: no sense in installing full /24 as the interface
* is point-to-point.
* * add "loopback" route towards 10.0.9.1 via V_loif, saving
* @ia ifp in the gateway and using @ia as an address source.
*
* 4) Adding address with or without prefix to loopback interface.
* Example: 192.0.2.1/24. Action:
* * add "prefix" host route via @ia interface, using @ia as an address source.
* Note: Skip installing /24 prefix as it would introduce TTL loop
* for the traffic destined to these addresses.
*/
/*
* Checks if @ia needs to install loopback route to @ia address via
* ifa_maintain_loopback_route().
*
* Return true on success.
*/
static bool
ia_need_loopback_route(const struct in_ifaddr *ia)
{
struct ifnet *ifp = ia->ia_ifp;
/* Case 4: Skip loopback interfaces */
if ((ifp->if_flags & IFF_LOOPBACK) ||
(ia->ia_addr.sin_addr.s_addr == INADDR_ANY))
return (false);
/* Clash avoidance: Skip p2p interfaces with both addresses are equal */
if ((ifp->if_flags & IFF_POINTOPOINT) &&
ia->ia_dstaddr.sin_addr.s_addr == ia->ia_addr.sin_addr.s_addr)
return (false);
/* Case 2: skip /32 prefixes */
if (!(ifp->if_flags & IFF_POINTOPOINT) &&
(ia->ia_sockmask.sin_addr.s_addr == INADDR_BROADCAST))
return (false);
return (true);
}
/*
* Calculate "prefix" route corresponding to @ia.
*/
static void
ia_getrtprefix(const struct in_ifaddr *ia, struct in_addr *prefix, struct in_addr *mask)
{
if (ia->ia_ifp->if_flags & IFF_POINTOPOINT) {
/* Case 3: return host route for dstaddr */
*prefix = ia->ia_dstaddr.sin_addr;
mask->s_addr = INADDR_BROADCAST;
} else if (ia->ia_ifp->if_flags & IFF_LOOPBACK) {
/* Case 4: return host route for ifaddr */
*prefix = ia->ia_addr.sin_addr;
mask->s_addr = INADDR_BROADCAST;
} else {
/* Cases 1,2: return actual ia prefix */
*prefix = ia->ia_addr.sin_addr;
*mask = ia->ia_sockmask.sin_addr;
prefix->s_addr &= mask->s_addr;
}
}
/*
* Adds or delete interface "prefix" route corresponding to @ifa.
* Returns 0 on success or errno.
*/
int
in_handle_ifaddr_route(int cmd, struct in_ifaddr *ia)
{
struct ifaddr *ifa = &ia->ia_ifa;
struct in_addr daddr, maddr;
struct sockaddr_in *pmask;
struct epoch_tracker et;
int error;
ia_getrtprefix(ia, &daddr, &maddr);
struct sockaddr_in mask = {
.sin_family = AF_INET,
.sin_len = sizeof(struct sockaddr_in),
.sin_addr = maddr,
};
pmask = (maddr.s_addr != INADDR_BROADCAST) ? &mask : NULL;
struct sockaddr_in dst = {
.sin_family = AF_INET,
.sin_len = sizeof(struct sockaddr_in),
.sin_addr.s_addr = daddr.s_addr & maddr.s_addr,
};
struct ifnet *ifp = ia->ia_ifp;
if ((maddr.s_addr == INADDR_BROADCAST) &&
(!(ia->ia_ifp->if_flags & (IFF_POINTOPOINT|IFF_LOOPBACK)))) {
/* Case 2: host route on broadcast interface */
ifp = V_loif;
}
uint32_t fibnum = ifa->ifa_ifp->if_fib;
NET_EPOCH_ENTER(et);
error = in_handle_prefix_route(fibnum, cmd, &dst, pmask, ifa, ifp);
NET_EPOCH_EXIT(et);
return (error);
}
/*
* Check if we have a route for the given prefix already.
*/
static bool
in_hasrtprefix(struct in_ifaddr *target)
{
struct epoch_tracker et;
struct in_ifaddr *ia;
struct in_addr prefix, mask, p, m;
bool result = false;
ia_getrtprefix(target, &prefix, &mask);
/* Look for an existing address with the same prefix, mask, and fib */
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
ia_getrtprefix(ia, &p, &m);
if (prefix.s_addr != p.s_addr ||
mask.s_addr != m.s_addr)
continue;
if (target->ia_ifp->if_fib != ia->ia_ifp->if_fib)
continue;
/*
* If we got a matching prefix route inserted by other
* interface address, we are done here.
*/
if (ia->ia_flags & IFA_ROUTE) {
result = true;
break;
}
}
NET_EPOCH_EXIT(et);
return (result);
}
int
in_addprefix(struct in_ifaddr *target)
{
int error;
if (in_hasrtprefix(target)) {
if (V_nosameprefix)
return (EEXIST);
else {
rt_addrmsg(RTM_ADD, &target->ia_ifa,
target->ia_ifp->if_fib);
return (0);
}
}
/*
* No-one seem to have this prefix route, so we try to insert it.
*/
rt_addrmsg(RTM_ADD, &target->ia_ifa, target->ia_ifp->if_fib);
error = in_handle_ifaddr_route(RTM_ADD, target);
if (!error)
target->ia_flags |= IFA_ROUTE;
return (error);
}
/*
* Removes either all lle entries for given @ia, or lle
* corresponding to @ia address.
*/
static void
in_scrubprefixlle(struct in_ifaddr *ia, int all, u_int flags)
{
struct sockaddr_in addr, mask;
struct sockaddr *saddr, *smask;
struct ifnet *ifp;
saddr = (struct sockaddr *)&addr;
bzero(&addr, sizeof(addr));
addr.sin_len = sizeof(addr);
addr.sin_family = AF_INET;
smask = (struct sockaddr *)&mask;
bzero(&mask, sizeof(mask));
mask.sin_len = sizeof(mask);
mask.sin_family = AF_INET;
mask.sin_addr.s_addr = ia->ia_subnetmask;
ifp = ia->ia_ifp;
if (all) {
/*
* Remove all L2 entries matching given prefix.
* Convert address to host representation to avoid
* doing this on every callback. ia_subnetmask is already
* stored in host representation.
*/
addr.sin_addr.s_addr = ntohl(ia->ia_addr.sin_addr.s_addr);
lltable_prefix_free(AF_INET, saddr, smask, flags);
} else {
/* Remove interface address only */
addr.sin_addr.s_addr = ia->ia_addr.sin_addr.s_addr;
lltable_delete_addr(LLTABLE(ifp), LLE_IFADDR, saddr);
}
}
/*
* If there is no other address in the system that can serve a route to the
* same prefix, remove the route. Hand over the route to the new address
* otherwise.
*/
int
in_scrubprefix(struct in_ifaddr *target, u_int flags)
{
struct epoch_tracker et;
struct in_ifaddr *ia;
struct in_addr prefix, mask, p, m;
int error = 0;
/*
* Remove the loopback route to the interface address.
*/
if (ia_need_loopback_route(target) && (flags & LLE_STATIC)) {
struct in_ifaddr *eia;
eia = in_localip_more(target);
if (eia != NULL) {
error = ifa_switch_loopback_route((struct ifaddr *)eia,
(struct sockaddr *)&target->ia_addr);
ifa_free(&eia->ia_ifa);
} else {
error = ifa_del_loopback_route((struct ifaddr *)target,
(struct sockaddr *)&target->ia_addr);
}
}
ia_getrtprefix(target, &prefix, &mask);
if ((target->ia_flags & IFA_ROUTE) == 0) {
rt_addrmsg(RTM_DELETE, &target->ia_ifa, target->ia_ifp->if_fib);
/*
* Removing address from !IFF_UP interface or
* prefix which exists on other interface (along with route).
* No entries should exist here except target addr.
* Given that, delete this entry only.
*/
in_scrubprefixlle(target, 0, flags);
return (0);
}
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
ia_getrtprefix(ia, &p, &m);
if (prefix.s_addr != p.s_addr ||
mask.s_addr != m.s_addr)
continue;
if ((ia->ia_ifp->if_flags & IFF_UP) == 0)
continue;
/*
* If we got a matching prefix address, move IFA_ROUTE and
* the route itself to it. Make sure that routing daemons
* get a heads-up.
*/
if ((ia->ia_flags & IFA_ROUTE) == 0) {
ifa_ref(&ia->ia_ifa);
NET_EPOCH_EXIT(et);
error = in_handle_ifaddr_route(RTM_DELETE, target);
if (error == 0)
target->ia_flags &= ~IFA_ROUTE;
else
log(LOG_INFO, "in_scrubprefix: err=%d, old prefix delete failed\n",
error);
/* Scrub all entries IFF interface is different */
in_scrubprefixlle(target, target->ia_ifp != ia->ia_ifp,
flags);
error = in_handle_ifaddr_route(RTM_ADD, ia);
if (error == 0)
ia->ia_flags |= IFA_ROUTE;
else
log(LOG_INFO, "in_scrubprefix: err=%d, new prefix add failed\n",
error);
ifa_free(&ia->ia_ifa);
return (error);
}
}
NET_EPOCH_EXIT(et);
/*
* remove all L2 entries on the given prefix
*/
in_scrubprefixlle(target, 1, flags);
/*
* As no-one seem to have this prefix, we can remove the route.
*/
rt_addrmsg(RTM_DELETE, &target->ia_ifa, target->ia_ifp->if_fib);
error = in_handle_ifaddr_route(RTM_DELETE, target);
if (error == 0)
target->ia_flags &= ~IFA_ROUTE;
else
log(LOG_INFO, "in_scrubprefix: err=%d, prefix delete failed\n", error);
return (error);
}
void
in_ifscrub_all(void)
{
struct ifnet *ifp;
struct ifaddr *ifa, *nifa;
struct ifaliasreq ifr;
IFNET_RLOCK();
CK_STAILQ_FOREACH(ifp, &V_ifnet, if_link) {
/* Cannot lock here - lock recursion. */
/* NET_EPOCH_ENTER(et); */
CK_STAILQ_FOREACH_SAFE(ifa, &ifp->if_addrhead, ifa_link, nifa) {
if (ifa->ifa_addr->sa_family != AF_INET)
continue;
/*
* This is ugly but the only way for legacy IP to
* cleanly remove addresses and everything attached.
*/
bzero(&ifr, sizeof(ifr));
ifr.ifra_addr = *ifa->ifa_addr;
if (ifa->ifa_dstaddr)
ifr.ifra_broadaddr = *ifa->ifa_dstaddr;
(void)in_control(NULL, SIOCDIFADDR, (caddr_t)&ifr,
ifp, NULL);
}
/* NET_EPOCH_EXIT(et); */
in_purgemaddrs(ifp);
igmp_domifdetach(ifp);
}
IFNET_RUNLOCK();
}
int
in_ifaddr_broadcast(struct in_addr in, struct in_ifaddr *ia)
{
return ((in.s_addr == ia->ia_broadaddr.sin_addr.s_addr ||
/*
* Optionally check for old-style (host 0) broadcast, but
* taking into account that RFC 3021 obsoletes it.
*/
(V_broadcast_lowest && ia->ia_subnetmask != IN_RFC3021_MASK &&
ntohl(in.s_addr) == ia->ia_subnet)) &&
/*
* Check for an all one subnetmask. These
* only exist when an interface gets a secondary
* address.
*/
ia->ia_subnetmask != (u_long)0xffffffff);
}
/*
* Return 1 if the address might be a local broadcast address.
*/
int
in_broadcast(struct in_addr in, struct ifnet *ifp)
{
struct ifaddr *ifa;
int found;
NET_EPOCH_ASSERT();
if (in.s_addr == INADDR_BROADCAST ||
in.s_addr == INADDR_ANY)
return (1);
if ((ifp->if_flags & IFF_BROADCAST) == 0)
return (0);
found = 0;
/*
* Look through the list of addresses for a match
* with a broadcast address.
*/
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
if (ifa->ifa_addr->sa_family == AF_INET &&
in_ifaddr_broadcast(in, (struct in_ifaddr *)ifa)) {
found = 1;
break;
}
return (found);
}
/*
* On interface removal, clean up IPv4 data structures hung off of the ifnet.
*/
void
in_ifdetach(struct ifnet *ifp)
{
IN_MULTI_LOCK();
in_pcbpurgeif0(&V_ripcbinfo, ifp);
in_pcbpurgeif0(&V_udbinfo, ifp);
in_pcbpurgeif0(&V_ulitecbinfo, ifp);
in_purgemaddrs(ifp);
IN_MULTI_UNLOCK();
/*
* Make sure all multicast deletions invoking if_ioctl() are
* completed before returning. Else we risk accessing a freed
* ifnet structure pointer.
*/
inm_release_wait(NULL);
}
/*
* Delete all IPv4 multicast address records, and associated link-layer
* multicast address records, associated with ifp.
* XXX It looks like domifdetach runs AFTER the link layer cleanup.
* XXX This should not race with ifma_protospec being set during
* a new allocation, if it does, we have bigger problems.
*/
static void
in_purgemaddrs(struct ifnet *ifp)
{
struct in_multi_head purgeinms;
struct in_multi *inm;
struct ifmultiaddr *ifma, *next;
SLIST_INIT(&purgeinms);
IN_MULTI_LIST_LOCK();
/*
* Extract list of in_multi associated with the detaching ifp
* which the PF_INET layer is about to release.
* We need to do this as IF_ADDR_LOCK() may be re-acquired
* by code further down.
*/
IF_ADDR_WLOCK(ifp);
restart:
CK_STAILQ_FOREACH_SAFE(ifma, &ifp->if_multiaddrs, ifma_link, next) {
if (ifma->ifma_addr->sa_family != AF_INET ||
ifma->ifma_protospec == NULL)
continue;
inm = (struct in_multi *)ifma->ifma_protospec;
inm_rele_locked(&purgeinms, inm);
if (__predict_false(ifma_restart)) {
ifma_restart = true;
goto restart;
}
}
IF_ADDR_WUNLOCK(ifp);
inm_release_list_deferred(&purgeinms);
igmp_ifdetach(ifp);
IN_MULTI_LIST_UNLOCK();
}
struct in_llentry {
struct llentry base;
};
#define IN_LLTBL_DEFAULT_HSIZE 32
#define IN_LLTBL_HASH(k, h) \
(((((((k >> 8) ^ k) >> 8) ^ k) >> 8) ^ k) & ((h) - 1))
/*
* Do actual deallocation of @lle.
*/
static void
in_lltable_destroy_lle_unlocked(epoch_context_t ctx)
{
struct llentry *lle;
lle = __containerof(ctx, struct llentry, lle_epoch_ctx);
LLE_LOCK_DESTROY(lle);
LLE_REQ_DESTROY(lle);
free(lle, M_LLTABLE);
}
/*
* Called by LLE_FREE_LOCKED when number of references
* drops to zero.
*/
static void
in_lltable_destroy_lle(struct llentry *lle)
{
LLE_WUNLOCK(lle);
NET_EPOCH_CALL(in_lltable_destroy_lle_unlocked, &lle->lle_epoch_ctx);
}
static struct llentry *
in_lltable_new(struct in_addr addr4, u_int flags)
{
struct in_llentry *lle;
lle = malloc(sizeof(struct in_llentry), M_LLTABLE, M_NOWAIT | M_ZERO);
if (lle == NULL) /* NB: caller generates msg */
return NULL;
/*
* For IPv4 this will trigger "arpresolve" to generate
* an ARP request.
*/
lle->base.la_expire = time_uptime; /* mark expired */
lle->base.r_l3addr.addr4 = addr4;
lle->base.lle_refcnt = 1;
lle->base.lle_free = in_lltable_destroy_lle;
LLE_LOCK_INIT(&lle->base);
LLE_REQ_INIT(&lle->base);
callout_init(&lle->base.lle_timer, 1);
return (&lle->base);
}
#define IN_ARE_MASKED_ADDR_EQUAL(d, a, m) ( \
((((d).s_addr ^ (a).s_addr) & (m).s_addr)) == 0 )
static int
in_lltable_match_prefix(const struct sockaddr *saddr,
const struct sockaddr *smask, u_int flags, struct llentry *lle)
{
struct in_addr addr, mask, lle_addr;
addr = ((const struct sockaddr_in *)saddr)->sin_addr;
mask = ((const struct sockaddr_in *)smask)->sin_addr;
lle_addr.s_addr = ntohl(lle->r_l3addr.addr4.s_addr);
if (IN_ARE_MASKED_ADDR_EQUAL(lle_addr, addr, mask) == 0)
return (0);
if (lle->la_flags & LLE_IFADDR) {
/*
* Delete LLE_IFADDR records IFF address & flag matches.
* Note that addr is the interface address within prefix
* being matched.
* Note also we should handle 'ifdown' cases without removing
* ifaddr macs.
*/
if (addr.s_addr == lle_addr.s_addr && (flags & LLE_STATIC) != 0)
return (1);
return (0);
}
/* flags & LLE_STATIC means deleting both dynamic and static entries */
if ((flags & LLE_STATIC) || !(lle->la_flags & LLE_STATIC))
return (1);
return (0);
}
static void
in_lltable_free_entry(struct lltable *llt, struct llentry *lle)
{
size_t pkts_dropped;
LLE_WLOCK_ASSERT(lle);
KASSERT(llt != NULL, ("lltable is NULL"));
/* Unlink entry from table if not already */
if ((lle->la_flags & LLE_LINKED) != 0) {
IF_AFDATA_WLOCK_ASSERT(llt->llt_ifp);
lltable_unlink_entry(llt, lle);
}
/* Drop hold queue */
pkts_dropped = llentry_free(lle);
ARPSTAT_ADD(dropped, pkts_dropped);
}
static int
in_lltable_rtcheck(struct ifnet *ifp, u_int flags, const struct sockaddr *l3addr)
{
struct nhop_object *nh;
struct in_addr addr;
KASSERT(l3addr->sa_family == AF_INET,
("sin_family %d", l3addr->sa_family));
addr = ((const struct sockaddr_in *)l3addr)->sin_addr;
nh = fib4_lookup(ifp->if_fib, addr, 0, NHR_NONE, 0);
if (nh == NULL)
return (EINVAL);
/*
* If the gateway for an existing host route matches the target L3
* address, which is a special route inserted by some implementation
* such as MANET, and the interface is of the correct type, then
* allow for ARP to proceed.
*/
if (nh->nh_flags & NHF_GATEWAY) {
if (!(nh->nh_flags & NHF_HOST) || nh->nh_ifp->if_type != IFT_ETHER ||
(nh->nh_ifp->if_flags & (IFF_NOARP | IFF_STATICARP)) != 0 ||
memcmp(nh->gw_sa.sa_data, l3addr->sa_data,
sizeof(in_addr_t)) != 0) {
return (EINVAL);
}
}
/*
* Make sure that at least the destination address is covered
* by the route. This is for handling the case where 2 or more
* interfaces have the same prefix. An incoming packet arrives
* on one interface and the corresponding outgoing packet leaves
* another interface.
*/
if ((nh->nh_ifp != ifp) && (nh->nh_flags & NHF_HOST) == 0) {
struct in_ifaddr *ia = (struct in_ifaddr *)ifaof_ifpforaddr(l3addr, ifp);
struct in_addr dst_addr, mask_addr;
if (ia == NULL)
return (EINVAL);
/*
* ifaof_ifpforaddr() returns _best matching_ IFA.
* It is possible that ifa prefix does not cover our address.
* Explicitly verify and fail if that's the case.
*/
dst_addr = IA_SIN(ia)->sin_addr;
mask_addr.s_addr = htonl(ia->ia_subnetmask);
if (!IN_ARE_MASKED_ADDR_EQUAL(dst_addr, addr, mask_addr))
return (EINVAL);
}
return (0);
}
static inline uint32_t
in_lltable_hash_dst(const struct in_addr dst, uint32_t hsize)
{
return (IN_LLTBL_HASH(dst.s_addr, hsize));
}
static uint32_t
in_lltable_hash(const struct llentry *lle, uint32_t hsize)
{
return (in_lltable_hash_dst(lle->r_l3addr.addr4, hsize));
}
static void
in_lltable_fill_sa_entry(const struct llentry *lle, struct sockaddr *sa)
{
struct sockaddr_in *sin;
sin = (struct sockaddr_in *)sa;
bzero(sin, sizeof(*sin));
sin->sin_family = AF_INET;
sin->sin_len = sizeof(*sin);
sin->sin_addr = lle->r_l3addr.addr4;
}
static inline struct llentry *
in_lltable_find_dst(struct lltable *llt, struct in_addr dst)
{
struct llentry *lle;
struct llentries *lleh;
u_int hashidx;
hashidx = in_lltable_hash_dst(dst, llt->llt_hsize);
lleh = &llt->lle_head[hashidx];
CK_LIST_FOREACH(lle, lleh, lle_next) {
if (lle->la_flags & LLE_DELETED)
continue;
if (lle->r_l3addr.addr4.s_addr == dst.s_addr)
break;
}
return (lle);
}
static void
in_lltable_delete_entry(struct lltable *llt, struct llentry *lle)
{
lle->la_flags |= LLE_DELETED;
EVENTHANDLER_INVOKE(lle_event, lle, LLENTRY_DELETED);
#ifdef DIAGNOSTIC
log(LOG_INFO, "ifaddr cache = %p is deleted\n", lle);
#endif
llentry_free(lle);
}
static struct llentry *
in_lltable_alloc(struct lltable *llt, u_int flags, const struct sockaddr *l3addr)
{
const struct sockaddr_in *sin = (const struct sockaddr_in *)l3addr;
struct ifnet *ifp = llt->llt_ifp;
struct llentry *lle;
char linkhdr[LLE_MAX_LINKHDR];
size_t linkhdrsize;
int lladdr_off;
KASSERT(l3addr->sa_family == AF_INET,
("sin_family %d", l3addr->sa_family));
/*
* A route that covers the given address must have
* been installed 1st because we are doing a resolution,
* verify this.
*/
if (!(flags & LLE_IFADDR) &&
in_lltable_rtcheck(ifp, flags, l3addr) != 0)
return (NULL);
lle = in_lltable_new(sin->sin_addr, flags);
if (lle == NULL) {
log(LOG_INFO, "lla_lookup: new lle malloc failed\n");
return (NULL);
}
lle->la_flags = flags;
if (flags & LLE_STATIC)
lle->r_flags |= RLLE_VALID;
if ((flags & LLE_IFADDR) == LLE_IFADDR) {
linkhdrsize = LLE_MAX_LINKHDR;
if (lltable_calc_llheader(ifp, AF_INET, IF_LLADDR(ifp),
linkhdr, &linkhdrsize, &lladdr_off) != 0) {
NET_EPOCH_CALL(in_lltable_destroy_lle_unlocked, &lle->lle_epoch_ctx);
return (NULL);
}
lltable_set_entry_addr(ifp, lle, linkhdr, linkhdrsize,
lladdr_off);
lle->la_flags |= LLE_STATIC;
lle->r_flags |= (RLLE_VALID | RLLE_IFADDR);
}
return (lle);
}
/*
* Return NULL if not found or marked for deletion.
* If found return lle read locked.
*/
static struct llentry *
in_lltable_lookup(struct lltable *llt, u_int flags, const struct sockaddr *l3addr)
{
const struct sockaddr_in *sin = (const struct sockaddr_in *)l3addr;
struct llentry *lle;
IF_AFDATA_LOCK_ASSERT(llt->llt_ifp);
KASSERT(l3addr->sa_family == AF_INET,
("sin_family %d", l3addr->sa_family));
KASSERT((flags & (LLE_UNLOCKED | LLE_EXCLUSIVE)) !=
(LLE_UNLOCKED | LLE_EXCLUSIVE),
("wrong lle request flags: %#x", flags));
lle = in_lltable_find_dst(llt, sin->sin_addr);
if (lle == NULL)
return (NULL);
if (flags & LLE_UNLOCKED)
return (lle);
if (flags & LLE_EXCLUSIVE)
LLE_WLOCK(lle);
else
LLE_RLOCK(lle);
/*
* If the afdata lock is not held, the LLE may have been unlinked while
* we were blocked on the LLE lock. Check for this case.
*/
if (__predict_false((lle->la_flags & LLE_LINKED) == 0)) {
if (flags & LLE_EXCLUSIVE)
LLE_WUNLOCK(lle);
else
LLE_RUNLOCK(lle);
return (NULL);
}
return (lle);
}
static int
in_lltable_dump_entry(struct lltable *llt, struct llentry *lle,
struct sysctl_req *wr)
{
struct ifnet *ifp = llt->llt_ifp;
/* XXX stack use */
struct {
struct rt_msghdr rtm;
struct sockaddr_in sin;
struct sockaddr_dl sdl;
} arpc;
struct sockaddr_dl *sdl;
int error;
bzero(&arpc, sizeof(arpc));
/* skip deleted entries */
if ((lle->la_flags & LLE_DELETED) == LLE_DELETED)
return (0);
/* Skip if jailed and not a valid IP of the prison. */
lltable_fill_sa_entry(lle,(struct sockaddr *)&arpc.sin);
if (prison_if(wr->td->td_ucred, (struct sockaddr *)&arpc.sin) != 0)
return (0);
/*
* produce a msg made of:
* struct rt_msghdr;
* struct sockaddr_in; (IPv4)
* struct sockaddr_dl;
*/
arpc.rtm.rtm_msglen = sizeof(arpc);
arpc.rtm.rtm_version = RTM_VERSION;
arpc.rtm.rtm_type = RTM_GET;
arpc.rtm.rtm_flags = RTF_UP;
arpc.rtm.rtm_addrs = RTA_DST | RTA_GATEWAY;
/* publish */
if (lle->la_flags & LLE_PUB)
arpc.rtm.rtm_flags |= RTF_ANNOUNCE;
sdl = &arpc.sdl;
sdl->sdl_family = AF_LINK;
sdl->sdl_len = sizeof(*sdl);
sdl->sdl_index = ifp->if_index;
sdl->sdl_type = ifp->if_type;
if ((lle->la_flags & LLE_VALID) == LLE_VALID) {
sdl->sdl_alen = ifp->if_addrlen;
bcopy(lle->ll_addr, LLADDR(sdl), ifp->if_addrlen);
} else {
sdl->sdl_alen = 0;
bzero(LLADDR(sdl), ifp->if_addrlen);
}
arpc.rtm.rtm_rmx.rmx_expire =
lle->la_flags & LLE_STATIC ? 0 : lle->la_expire;
arpc.rtm.rtm_flags |= (RTF_HOST | RTF_LLDATA);
if (lle->la_flags & LLE_STATIC)
arpc.rtm.rtm_flags |= RTF_STATIC;
if (lle->la_flags & LLE_IFADDR)
arpc.rtm.rtm_flags |= RTF_PINNED;
arpc.rtm.rtm_index = ifp->if_index;
error = SYSCTL_OUT(wr, &arpc, sizeof(arpc));
return (error);
}
static struct lltable *
in_lltattach(struct ifnet *ifp)
{
struct lltable *llt;
llt = lltable_allocate_htbl(IN_LLTBL_DEFAULT_HSIZE);
llt->llt_af = AF_INET;
llt->llt_ifp = ifp;
llt->llt_lookup = in_lltable_lookup;
llt->llt_alloc_entry = in_lltable_alloc;
llt->llt_delete_entry = in_lltable_delete_entry;
llt->llt_dump_entry = in_lltable_dump_entry;
llt->llt_hash = in_lltable_hash;
llt->llt_fill_sa_entry = in_lltable_fill_sa_entry;
llt->llt_free_entry = in_lltable_free_entry;
llt->llt_match_prefix = in_lltable_match_prefix;
llt->llt_mark_used = llentry_mark_used;
lltable_link(llt);
return (llt);
}
void *
in_domifattach(struct ifnet *ifp)
{
struct in_ifinfo *ii;
ii = malloc(sizeof(struct in_ifinfo), M_IFADDR, M_WAITOK|M_ZERO);
ii->ii_llt = in_lltattach(ifp);
ii->ii_igmp = igmp_domifattach(ifp);
return (ii);
}
void
in_domifdetach(struct ifnet *ifp, void *aux)
{
struct in_ifinfo *ii = (struct in_ifinfo *)aux;
igmp_domifdetach(ifp);
lltable_free(ii->ii_llt);
free(ii, M_IFADDR);
}
diff --git a/sys/netinet/in.h b/sys/netinet/in.h
index 92d150701422..1fc5c173b33c 100644
--- a/sys/netinet/in.h
+++ b/sys/netinet/in.h
@@ -1,689 +1,690 @@
/*-
* SPDX-License-Identifier: BSD-3-Clause
*
* Copyright (c) 1982, 1986, 1990, 1993
* The Regents of the University of California. 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 University 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 REGENTS 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 REGENTS 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.
*
* @(#)in.h 8.3 (Berkeley) 1/3/94
* $FreeBSD$
*/
#ifndef _NETINET_IN_H_
#define _NETINET_IN_H_
#include <sys/cdefs.h>
#include <sys/_types.h>
#include <machine/endian.h>
/* Protocols common to RFC 1700, POSIX, and X/Open. */
#define IPPROTO_IP 0 /* dummy for IP */
#define IPPROTO_ICMP 1 /* control message protocol */
#define IPPROTO_TCP 6 /* tcp */
#define IPPROTO_UDP 17 /* user datagram protocol */
#define INADDR_ANY ((in_addr_t)0x00000000)
#define INADDR_BROADCAST ((in_addr_t)0xffffffff) /* must be masked */
#ifndef _UINT8_T_DECLARED
typedef __uint8_t uint8_t;
#define _UINT8_T_DECLARED
#endif
#ifndef _UINT16_T_DECLARED
typedef __uint16_t uint16_t;
#define _UINT16_T_DECLARED
#endif
#ifndef _UINT32_T_DECLARED
typedef __uint32_t uint32_t;
#define _UINT32_T_DECLARED
#endif
#ifndef _IN_ADDR_T_DECLARED
typedef uint32_t in_addr_t;
#define _IN_ADDR_T_DECLARED
#endif
#ifndef _IN_PORT_T_DECLARED
typedef uint16_t in_port_t;
#define _IN_PORT_T_DECLARED
#endif
#ifndef _SA_FAMILY_T_DECLARED
typedef __sa_family_t sa_family_t;
#define _SA_FAMILY_T_DECLARED
#endif
/* Internet address (a structure for historical reasons). */
#ifndef _STRUCT_IN_ADDR_DECLARED
struct in_addr {
in_addr_t s_addr;
};
#define _STRUCT_IN_ADDR_DECLARED
#endif
#ifndef _SOCKLEN_T_DECLARED
typedef __socklen_t socklen_t;
#define _SOCKLEN_T_DECLARED
#endif
#include <sys/_sockaddr_storage.h>
/* Socket address, internet style. */
struct sockaddr_in {
uint8_t sin_len;
sa_family_t sin_family;
in_port_t sin_port;
struct in_addr sin_addr;
char sin_zero[8];
};
#if !defined(_KERNEL) && __POSIX_VISIBLE >= 200112
#ifndef _BYTEORDER_PROTOTYPED
#define _BYTEORDER_PROTOTYPED
__BEGIN_DECLS
uint32_t htonl(uint32_t);
uint16_t htons(uint16_t);
uint32_t ntohl(uint32_t);
uint16_t ntohs(uint16_t);
__END_DECLS
#endif
#ifndef _BYTEORDER_FUNC_DEFINED
#define _BYTEORDER_FUNC_DEFINED
#define htonl(x) __htonl(x)
#define htons(x) __htons(x)
#define ntohl(x) __ntohl(x)
#define ntohs(x) __ntohs(x)
#endif
#endif /* !_KERNEL && __POSIX_VISIBLE >= 200112 */
#if __POSIX_VISIBLE >= 200112
#define IPPROTO_IPV6 41 /* IP6 header */
#define IPPROTO_RAW 255 /* raw IP packet */
#define INET_ADDRSTRLEN 16
#endif
#if __BSD_VISIBLE
/*
* Constants and structures defined by the internet system,
* Per RFC 790, September 1981, and numerous additions.
*/
/*
* Protocols (RFC 1700)
*/
#define IPPROTO_HOPOPTS 0 /* IP6 hop-by-hop options */
#define IPPROTO_IGMP 2 /* group mgmt protocol */
#define IPPROTO_GGP 3 /* gateway^2 (deprecated) */
#define IPPROTO_IPV4 4 /* IPv4 encapsulation */
#define IPPROTO_IPIP IPPROTO_IPV4 /* for compatibility */
#define IPPROTO_ST 7 /* Stream protocol II */
#define IPPROTO_EGP 8 /* exterior gateway protocol */
#define IPPROTO_PIGP 9 /* private interior gateway */
#define IPPROTO_RCCMON 10 /* BBN RCC Monitoring */
#define IPPROTO_NVPII 11 /* network voice protocol*/
#define IPPROTO_PUP 12 /* pup */
#define IPPROTO_ARGUS 13 /* Argus */
#define IPPROTO_EMCON 14 /* EMCON */
#define IPPROTO_XNET 15 /* Cross Net Debugger */
#define IPPROTO_CHAOS 16 /* Chaos*/
#define IPPROTO_MUX 18 /* Multiplexing */
#define IPPROTO_MEAS 19 /* DCN Measurement Subsystems */
#define IPPROTO_HMP 20 /* Host Monitoring */
#define IPPROTO_PRM 21 /* Packet Radio Measurement */
#define IPPROTO_IDP 22 /* xns idp */
#define IPPROTO_TRUNK1 23 /* Trunk-1 */
#define IPPROTO_TRUNK2 24 /* Trunk-2 */
#define IPPROTO_LEAF1 25 /* Leaf-1 */
#define IPPROTO_LEAF2 26 /* Leaf-2 */
#define IPPROTO_RDP 27 /* Reliable Data */
#define IPPROTO_IRTP 28 /* Reliable Transaction */
#define IPPROTO_TP 29 /* tp-4 w/ class negotiation */
#define IPPROTO_BLT 30 /* Bulk Data Transfer */
#define IPPROTO_NSP 31 /* Network Services */
#define IPPROTO_INP 32 /* Merit Internodal */
#define IPPROTO_DCCP 33 /* Datagram Congestion Control Protocol */
#define IPPROTO_3PC 34 /* Third Party Connect */
#define IPPROTO_IDPR 35 /* InterDomain Policy Routing */
#define IPPROTO_XTP 36 /* XTP */
#define IPPROTO_DDP 37 /* Datagram Delivery */
#define IPPROTO_CMTP 38 /* Control Message Transport */
#define IPPROTO_TPXX 39 /* TP++ Transport */
#define IPPROTO_IL 40 /* IL transport protocol */
#define IPPROTO_SDRP 42 /* Source Demand Routing */
#define IPPROTO_ROUTING 43 /* IP6 routing header */
#define IPPROTO_FRAGMENT 44 /* IP6 fragmentation header */
#define IPPROTO_IDRP 45 /* InterDomain Routing*/
#define IPPROTO_RSVP 46 /* resource reservation */
#define IPPROTO_GRE 47 /* General Routing Encap. */
#define IPPROTO_MHRP 48 /* Mobile Host Routing */
#define IPPROTO_BHA 49 /* BHA */
#define IPPROTO_ESP 50 /* IP6 Encap Sec. Payload */
#define IPPROTO_AH 51 /* IP6 Auth Header */
#define IPPROTO_INLSP 52 /* Integ. Net Layer Security */
#define IPPROTO_SWIPE 53 /* IP with encryption */
#define IPPROTO_NHRP 54 /* Next Hop Resolution */
#define IPPROTO_MOBILE 55 /* IP Mobility */
#define IPPROTO_TLSP 56 /* Transport Layer Security */
#define IPPROTO_SKIP 57 /* SKIP */
#define IPPROTO_ICMPV6 58 /* ICMP6 */
#define IPPROTO_NONE 59 /* IP6 no next header */
#define IPPROTO_DSTOPTS 60 /* IP6 destination option */
#define IPPROTO_AHIP 61 /* any host internal protocol */
#define IPPROTO_CFTP 62 /* CFTP */
#define IPPROTO_HELLO 63 /* "hello" routing protocol */
#define IPPROTO_SATEXPAK 64 /* SATNET/Backroom EXPAK */
#define IPPROTO_KRYPTOLAN 65 /* Kryptolan */
#define IPPROTO_RVD 66 /* Remote Virtual Disk */
#define IPPROTO_IPPC 67 /* Pluribus Packet Core */
#define IPPROTO_ADFS 68 /* Any distributed FS */
#define IPPROTO_SATMON 69 /* Satnet Monitoring */
#define IPPROTO_VISA 70 /* VISA Protocol */
#define IPPROTO_IPCV 71 /* Packet Core Utility */
#define IPPROTO_CPNX 72 /* Comp. Prot. Net. Executive */
#define IPPROTO_CPHB 73 /* Comp. Prot. HeartBeat */
#define IPPROTO_WSN 74 /* Wang Span Network */
#define IPPROTO_PVP 75 /* Packet Video Protocol */
#define IPPROTO_BRSATMON 76 /* BackRoom SATNET Monitoring */
#define IPPROTO_ND 77 /* Sun net disk proto (temp.) */
#define IPPROTO_WBMON 78 /* WIDEBAND Monitoring */
#define IPPROTO_WBEXPAK 79 /* WIDEBAND EXPAK */
#define IPPROTO_EON 80 /* ISO cnlp */
#define IPPROTO_VMTP 81 /* VMTP */
#define IPPROTO_SVMTP 82 /* Secure VMTP */
#define IPPROTO_VINES 83 /* Banyon VINES */
#define IPPROTO_TTP 84 /* TTP */
#define IPPROTO_IGP 85 /* NSFNET-IGP */
#define IPPROTO_DGP 86 /* dissimilar gateway prot. */
#define IPPROTO_TCF 87 /* TCF */
#define IPPROTO_IGRP 88 /* Cisco/GXS IGRP */
#define IPPROTO_OSPFIGP 89 /* OSPFIGP */
#define IPPROTO_SRPC 90 /* Strite RPC protocol */
#define IPPROTO_LARP 91 /* Locus Address Resoloution */
#define IPPROTO_MTP 92 /* Multicast Transport */
#define IPPROTO_AX25 93 /* AX.25 Frames */
#define IPPROTO_IPEIP 94 /* IP encapsulated in IP */
#define IPPROTO_MICP 95 /* Mobile Int.ing control */
#define IPPROTO_SCCSP 96 /* Semaphore Comm. security */
#define IPPROTO_ETHERIP 97 /* Ethernet IP encapsulation */
#define IPPROTO_ENCAP 98 /* encapsulation header */
#define IPPROTO_APES 99 /* any private encr. scheme */
#define IPPROTO_GMTP 100 /* GMTP*/
#define IPPROTO_IPCOMP 108 /* payload compression (IPComp) */
#define IPPROTO_SCTP 132 /* SCTP */
#define IPPROTO_MH 135 /* IPv6 Mobility Header */
#define IPPROTO_UDPLITE 136 /* UDP-Lite */
#define IPPROTO_HIP 139 /* IP6 Host Identity Protocol */
#define IPPROTO_SHIM6 140 /* IP6 Shim6 Protocol */
/* 101-254: Partly Unassigned */
#define IPPROTO_PIM 103 /* Protocol Independent Mcast */
#define IPPROTO_CARP 112 /* CARP */
#define IPPROTO_PGM 113 /* PGM */
#define IPPROTO_MPLS 137 /* MPLS-in-IP */
#define IPPROTO_PFSYNC 240 /* PFSYNC */
#define IPPROTO_RESERVED_253 253 /* Reserved */
#define IPPROTO_RESERVED_254 254 /* Reserved */
/* 255: Reserved */
/* BSD Private, local use, namespace incursion, no longer used */
#define IPPROTO_OLD_DIVERT 254 /* OLD divert pseudo-proto */
#define IPPROTO_MAX 256
/* last return value of *_input(), meaning "all job for this pkt is done". */
#define IPPROTO_DONE 257
/* Only used internally, so can be outside the range of valid IP protocols. */
#define IPPROTO_DIVERT 258 /* divert pseudo-protocol */
#define IPPROTO_SEND 259 /* SeND pseudo-protocol */
/*
* Defined to avoid confusion. The master value is defined by
* PROTO_SPACER in sys/protosw.h.
*/
#define IPPROTO_SPACER 32767 /* spacer for loadable protos */
/*
* Local port number conventions:
*
* When a user does a bind(2) or connect(2) with a port number of zero,
* a non-conflicting local port address is chosen.
* The default range is IPPORT_HIFIRSTAUTO through
* IPPORT_HILASTAUTO, although that is settable by sysctl.
*
* A user may set the IPPROTO_IP option IP_PORTRANGE to change this
* default assignment range.
*
* The value IP_PORTRANGE_DEFAULT causes the default behavior.
*
* The value IP_PORTRANGE_HIGH changes the range of candidate port numbers
* into the "high" range. These are reserved for client outbound connections
* which do not want to be filtered by any firewalls.
*
* The value IP_PORTRANGE_LOW changes the range to the "low" are
* that is (by convention) restricted to privileged processes. This
* convention is based on "vouchsafe" principles only. It is only secure
* if you trust the remote host to restrict these ports.
*
* The default range of ports and the high range can be changed by
* sysctl(3). (net.inet.ip.portrange.{hi,low,}{first,last})
*
* Changing those values has bad security implications if you are
* using a stateless firewall that is allowing packets outside of that
* range in order to allow transparent outgoing connections.
*
* Such a firewall configuration will generally depend on the use of these
* default values. If you change them, you may find your Security
* Administrator looking for you with a heavy object.
*
* For a slightly more orthodox text view on this:
*
* ftp://ftp.isi.edu/in-notes/iana/assignments/port-numbers
*
* port numbers are divided into three ranges:
*
* 0 - 1023 Well Known Ports
* 1024 - 49151 Registered Ports
* 49152 - 65535 Dynamic and/or Private Ports
*
*/
/*
* Ports < IPPORT_RESERVED are reserved for
* privileged processes (e.g. root). (IP_PORTRANGE_LOW)
*/
#define IPPORT_RESERVED 1024
/*
* Default local port range, used by IP_PORTRANGE_DEFAULT
*/
#define IPPORT_EPHEMERALFIRST 10000
#define IPPORT_EPHEMERALLAST 65535
/*
* Dynamic port range, used by IP_PORTRANGE_HIGH.
*/
#define IPPORT_HIFIRSTAUTO 49152
#define IPPORT_HILASTAUTO 65535
/*
* Scanning for a free reserved port return a value below IPPORT_RESERVED,
* but higher than IPPORT_RESERVEDSTART. Traditionally the start value was
* 512, but that conflicts with some well-known-services that firewalls may
* have a fit if we use.
*/
#define IPPORT_RESERVEDSTART 600
#define IPPORT_MAX 65535
/*
* Historical definitions of bits in internet address integers
* (pre-CIDR). Class A/B/C are long obsolete, and now deprecated.
* Hide these definitions from the kernel unless IN_HISTORICAL_NETS
* is defined. Provide the historical definitions to user level for now.
*/
#ifndef _KERNEL
#define IN_HISTORICAL_NETS
#endif
#ifdef IN_HISTORICAL_NETS
#define IN_CLASSA(i) (((in_addr_t)(i) & 0x80000000) == 0)
#define IN_CLASSA_NET 0xff000000
#define IN_CLASSA_NSHIFT 24
#define IN_CLASSA_HOST 0x00ffffff
#define IN_CLASSA_MAX 128
#define IN_CLASSB(i) (((in_addr_t)(i) & 0xc0000000) == 0x80000000)
#define IN_CLASSB_NET 0xffff0000
#define IN_CLASSB_NSHIFT 16
#define IN_CLASSB_HOST 0x0000ffff
#define IN_CLASSB_MAX 65536
#define IN_CLASSC(i) (((in_addr_t)(i) & 0xe0000000) == 0xc0000000)
#define IN_CLASSC_NET 0xffffff00
#define IN_CLASSC_NSHIFT 8
#define IN_CLASSC_HOST 0x000000ff
#endif /* IN_HISTORICAL_NETS */
#define IN_NETMASK_DEFAULT 0xffffff00 /* mask when forced to guess */
#define IN_MULTICAST(i) (((in_addr_t)(i) & 0xf0000000) == 0xe0000000)
#ifdef IN_HISTORICAL_NETS
#define IN_CLASSD(i) IN_MULTICAST(i)
#define IN_CLASSD_NET 0xf0000000 /* These ones aren't really */
#define IN_CLASSD_NSHIFT 28 /* net and host fields, but */
#define IN_CLASSD_HOST 0x0fffffff /* routing needn't know. */
#endif /* IN_HISTORICAL_NETS */
#define IN_EXPERIMENTAL(i) (((in_addr_t)(i) & 0xf0000000) == 0xf0000000)
#define IN_BADCLASS(i) (((in_addr_t)(i) & 0xf0000000) == 0xf0000000)
#define IN_LINKLOCAL(i) (((in_addr_t)(i) & 0xffff0000) == 0xa9fe0000)
#define IN_LOOPBACK(i) (((in_addr_t)(i) & 0xff000000) == 0x7f000000)
#define IN_ZERONET(i) (((in_addr_t)(i) & 0xff000000) == 0)
#define IN_PRIVATE(i) ((((in_addr_t)(i) & 0xff000000) == 0x0a000000) || \
(((in_addr_t)(i) & 0xfff00000) == 0xac100000) || \
(((in_addr_t)(i) & 0xffff0000) == 0xc0a80000))
#define IN_LOCAL_GROUP(i) (((in_addr_t)(i) & 0xffffff00) == 0xe0000000)
#define IN_ANY_LOCAL(i) (IN_LINKLOCAL(i) || IN_LOCAL_GROUP(i))
#define INADDR_LOOPBACK ((in_addr_t)0x7f000001)
#ifndef _KERNEL
#define INADDR_NONE ((in_addr_t)0xffffffff) /* -1 return */
#endif
#define INADDR_UNSPEC_GROUP ((in_addr_t)0xe0000000) /* 224.0.0.0 */
#define INADDR_ALLHOSTS_GROUP ((in_addr_t)0xe0000001) /* 224.0.0.1 */
#define INADDR_ALLRTRS_GROUP ((in_addr_t)0xe0000002) /* 224.0.0.2 */
#define INADDR_ALLRPTS_GROUP ((in_addr_t)0xe0000016) /* 224.0.0.22, IGMPv3 */
#define INADDR_CARP_GROUP ((in_addr_t)0xe0000012) /* 224.0.0.18 */
#define INADDR_PFSYNC_GROUP ((in_addr_t)0xe00000f0) /* 224.0.0.240 */
#define INADDR_ALLMDNS_GROUP ((in_addr_t)0xe00000fb) /* 224.0.0.251 */
#define INADDR_MAX_LOCAL_GROUP ((in_addr_t)0xe00000ff) /* 224.0.0.255 */
#ifdef IN_HISTORICAL_NETS
#define IN_LOOPBACKNET 127 /* official! */
#endif /* IN_HISTORICAL_NETS */
#define IN_RFC3021_MASK ((in_addr_t)0xfffffffe)
/*
* Options for use with [gs]etsockopt at the IP level.
* First word of comment is data type; bool is stored in int.
*/
#define IP_OPTIONS 1 /* buf/ip_opts; set/get IP options */
#define IP_HDRINCL 2 /* int; header is included with data */
#define IP_TOS 3 /* int; IP type of service and preced. */
#define IP_TTL 4 /* int; IP time to live */
#define IP_RECVOPTS 5 /* bool; receive all IP opts w/dgram */
#define IP_RECVRETOPTS 6 /* bool; receive IP opts for response */
#define IP_RECVDSTADDR 7 /* bool; receive IP dst addr w/dgram */
#define IP_SENDSRCADDR IP_RECVDSTADDR /* cmsg_type to set src addr */
#define IP_RETOPTS 8 /* ip_opts; set/get IP options */
#define IP_MULTICAST_IF 9 /* struct in_addr *or* struct ip_mreqn;
* set/get IP multicast i/f */
#define IP_MULTICAST_TTL 10 /* u_char; set/get IP multicast ttl */
#define IP_MULTICAST_LOOP 11 /* u_char; set/get IP multicast loopback */
#define IP_ADD_MEMBERSHIP 12 /* ip_mreq; add an IP group membership */
#define IP_DROP_MEMBERSHIP 13 /* ip_mreq; drop an IP group membership */
#define IP_MULTICAST_VIF 14 /* set/get IP mcast virt. iface */
#define IP_RSVP_ON 15 /* enable RSVP in kernel */
#define IP_RSVP_OFF 16 /* disable RSVP in kernel */
#define IP_RSVP_VIF_ON 17 /* set RSVP per-vif socket */
#define IP_RSVP_VIF_OFF 18 /* unset RSVP per-vif socket */
#define IP_PORTRANGE 19 /* int; range to choose for unspec port */
#define IP_RECVIF 20 /* bool; receive reception if w/dgram */
/* for IPSEC */
#define IP_IPSEC_POLICY 21 /* int; set/get security policy */
/* unused; was IP_FAITH */
#define IP_ONESBCAST 23 /* bool: send all-ones broadcast */
#define IP_BINDANY 24 /* bool: allow bind to any address */
#define IP_BINDMULTI 25 /* bool: allow multiple listeners on a tuple */
#define IP_RSS_LISTEN_BUCKET 26 /* int; set RSS listen bucket */
#define IP_ORIGDSTADDR 27 /* bool: receive IP dst addr/port w/dgram */
#define IP_RECVORIGDSTADDR IP_ORIGDSTADDR
/*
* Options for controlling the firewall and dummynet.
* Historical options (from 40 to 64) will eventually be
* replaced by only two options, IP_FW3 and IP_DUMMYNET3.
*/
#define IP_FW_TABLE_ADD 40 /* add entry */
#define IP_FW_TABLE_DEL 41 /* delete entry */
#define IP_FW_TABLE_FLUSH 42 /* flush table */
#define IP_FW_TABLE_GETSIZE 43 /* get table size */
#define IP_FW_TABLE_LIST 44 /* list table contents */
#define IP_FW3 48 /* generic ipfw v.3 sockopts */
#define IP_DUMMYNET3 49 /* generic dummynet v.3 sockopts */
#define IP_FW_ADD 50 /* add a firewall rule to chain */
#define IP_FW_DEL 51 /* delete a firewall rule from chain */
#define IP_FW_FLUSH 52 /* flush firewall rule chain */
#define IP_FW_ZERO 53 /* clear single/all firewall counter(s) */
#define IP_FW_GET 54 /* get entire firewall rule chain */
#define IP_FW_RESETLOG 55 /* reset logging counters */
#define IP_FW_NAT_CFG 56 /* add/config a nat rule */
#define IP_FW_NAT_DEL 57 /* delete a nat rule */
#define IP_FW_NAT_GET_CONFIG 58 /* get configuration of a nat rule */
#define IP_FW_NAT_GET_LOG 59 /* get log of a nat rule */
#define IP_DUMMYNET_CONFIGURE 60 /* add/configure a dummynet pipe */
#define IP_DUMMYNET_DEL 61 /* delete a dummynet pipe from chain */
#define IP_DUMMYNET_FLUSH 62 /* flush dummynet */
#define IP_DUMMYNET_GET 64 /* get entire dummynet pipes */
#define IP_RECVTTL 65 /* bool; receive IP TTL w/dgram */
#define IP_MINTTL 66 /* minimum TTL for packet or drop */
#define IP_DONTFRAG 67 /* don't fragment packet */
#define IP_RECVTOS 68 /* bool; receive IP TOS w/dgram */
/* IPv4 Source Filter Multicast API [RFC3678] */
#define IP_ADD_SOURCE_MEMBERSHIP 70 /* join a source-specific group */
#define IP_DROP_SOURCE_MEMBERSHIP 71 /* drop a single source */
#define IP_BLOCK_SOURCE 72 /* block a source */
#define IP_UNBLOCK_SOURCE 73 /* unblock a source */
/* The following option is private; do not use it from user applications. */
#define IP_MSFILTER 74 /* set/get filter list */
/* The following option deals with the 802.1Q Ethernet Priority Code Point */
#define IP_VLAN_PCP 75 /* int; set/get PCP used for packet, */
/* -1 use interface default */
/* Protocol Independent Multicast API [RFC3678] */
#define MCAST_JOIN_GROUP 80 /* join an any-source group */
#define MCAST_LEAVE_GROUP 81 /* leave all sources for group */
#define MCAST_JOIN_SOURCE_GROUP 82 /* join a source-specific group */
#define MCAST_LEAVE_SOURCE_GROUP 83 /* leave a single source */
#define MCAST_BLOCK_SOURCE 84 /* block a source */
#define MCAST_UNBLOCK_SOURCE 85 /* unblock a source */
/* Flow and RSS definitions */
#define IP_FLOWID 90 /* get flow id for the given socket/inp */
#define IP_FLOWTYPE 91 /* get flow type (M_HASHTYPE) */
#define IP_RSSBUCKETID 92 /* get RSS flowid -> bucket mapping */
#define IP_RECVFLOWID 93 /* bool; receive IP flowid/flowtype w/ datagram */
#define IP_RECVRSSBUCKETID 94 /* bool; receive IP RSS bucket id w/ datagram */
/*
* Defaults and limits for options
*/
#define IP_DEFAULT_MULTICAST_TTL 1 /* normally limit m'casts to 1 hop */
#define IP_DEFAULT_MULTICAST_LOOP 1 /* normally hear sends if a member */
/*
* Limit for IPv4 multicast memberships
*/
#define IP_MAX_MEMBERSHIPS 4095
/*
* Default resource limits for IPv4 multicast source filtering.
* These may be modified by sysctl.
*/
#define IP_MAX_GROUP_SRC_FILTER 512 /* sources per group */
#define IP_MAX_SOCK_SRC_FILTER 128 /* sources per socket/group */
#define IP_MAX_SOCK_MUTE_FILTER 128 /* XXX no longer used */
/*
* Argument structure for IP_ADD_MEMBERSHIP and IP_DROP_MEMBERSHIP.
*/
struct ip_mreq {
struct in_addr imr_multiaddr; /* IP multicast address of group */
struct in_addr imr_interface; /* local IP address of interface */
};
/*
* Modified argument structure for IP_MULTICAST_IF, obtained from Linux.
* This is used to specify an interface index for multicast sends, as
* the IPv4 legacy APIs do not support this (unless IP_SENDIF is available).
*/
struct ip_mreqn {
struct in_addr imr_multiaddr; /* IP multicast address of group */
struct in_addr imr_address; /* local IP address of interface */
int imr_ifindex; /* Interface index; cast to uint32_t */
};
/*
* Argument structure for IPv4 Multicast Source Filter APIs. [RFC3678]
*/
struct ip_mreq_source {
struct in_addr imr_multiaddr; /* IP multicast address of group */
struct in_addr imr_sourceaddr; /* IP address of source */
struct in_addr imr_interface; /* local IP address of interface */
};
/*
* Argument structures for Protocol-Independent Multicast Source
* Filter APIs. [RFC3678]
*/
struct group_req {
uint32_t gr_interface; /* interface index */
struct sockaddr_storage gr_group; /* group address */
};
struct group_source_req {
uint32_t gsr_interface; /* interface index */
struct sockaddr_storage gsr_group; /* group address */
struct sockaddr_storage gsr_source; /* source address */
};
#ifndef __MSFILTERREQ_DEFINED
#define __MSFILTERREQ_DEFINED
/*
* The following structure is private; do not use it from user applications.
* It is used to communicate IP_MSFILTER/IPV6_MSFILTER information between
* the RFC 3678 libc functions and the kernel.
*/
struct __msfilterreq {
uint32_t msfr_ifindex; /* interface index */
uint32_t msfr_fmode; /* filter mode for group */
uint32_t msfr_nsrcs; /* # of sources in msfr_srcs */
struct sockaddr_storage msfr_group; /* group address */
struct sockaddr_storage *msfr_srcs; /* pointer to the first member
* of a contiguous array of
* sources to filter in full.
*/
};
#endif
struct sockaddr;
/*
* Advanced (Full-state) APIs [RFC3678]
* The RFC specifies uint_t for the 6th argument to [sg]etsourcefilter().
* We use uint32_t here to be consistent.
*/
int setipv4sourcefilter(int, struct in_addr, struct in_addr, uint32_t,
uint32_t, struct in_addr *);
int getipv4sourcefilter(int, struct in_addr, struct in_addr, uint32_t *,
uint32_t *, struct in_addr *);
int setsourcefilter(int, uint32_t, struct sockaddr *, socklen_t,
uint32_t, uint32_t, struct sockaddr_storage *);
int getsourcefilter(int, uint32_t, struct sockaddr *, socklen_t,
uint32_t *, uint32_t *, struct sockaddr_storage *);
/*
* Filter modes; also used to represent per-socket filter mode internally.
*/
#define MCAST_UNDEFINED 0 /* fmode: not yet defined */
#define MCAST_INCLUDE 1 /* fmode: include these source(s) */
#define MCAST_EXCLUDE 2 /* fmode: exclude these source(s) */
/*
* Argument for IP_PORTRANGE:
* - which range to search when port is unspecified at bind() or connect()
*/
#define IP_PORTRANGE_DEFAULT 0 /* default range */
#define IP_PORTRANGE_HIGH 1 /* "high" - request firewall bypass */
#define IP_PORTRANGE_LOW 2 /* "low" - vouchsafe security */
/*
* Identifiers for IP sysctl nodes
*/
#define IPCTL_FORWARDING 1 /* act as router */
#define IPCTL_SENDREDIRECTS 2 /* may send redirects when forwarding */
#define IPCTL_DEFTTL 3 /* default TTL */
#ifdef notyet
#define IPCTL_DEFMTU 4 /* default MTU */
#endif
/* IPCTL_RTEXPIRE 5 deprecated */
/* IPCTL_RTMINEXPIRE 6 deprecated */
/* IPCTL_RTMAXCACHE 7 deprecated */
#define IPCTL_SOURCEROUTE 8 /* may perform source routes */
#define IPCTL_DIRECTEDBROADCAST 9 /* may re-broadcast received packets */
#define IPCTL_INTRQMAXLEN 10 /* max length of netisr queue */
#define IPCTL_INTRQDROPS 11 /* number of netisr q drops */
#define IPCTL_STATS 12 /* ipstat structure */
#define IPCTL_ACCEPTSOURCEROUTE 13 /* may accept source routed packets */
#define IPCTL_FASTFORWARDING 14 /* use fast IP forwarding code */
/* 15, unused, was: IPCTL_KEEPFAITH */
#define IPCTL_GIF_TTL 16 /* default TTL for gif encap packet */
#define IPCTL_INTRDQMAXLEN 17 /* max length of direct netisr queue */
#define IPCTL_INTRDQDROPS 18 /* number of direct netisr q drops */
#endif /* __BSD_VISIBLE */
#ifdef _KERNEL
struct ifnet; struct mbuf; /* forward declarations for Standard C */
struct in_ifaddr;
int in_broadcast(struct in_addr, struct ifnet *);
int in_ifaddr_broadcast(struct in_addr, struct in_ifaddr *);
int in_canforward(struct in_addr);
int in_localaddr(struct in_addr);
bool in_localip(struct in_addr);
+bool in_localip_fib(struct in_addr, uint16_t);
int in_ifhasaddr(struct ifnet *, struct in_addr);
struct in_ifaddr *in_findlocal(uint32_t, bool);
int inet_aton(const char *, struct in_addr *); /* in libkern */
char *inet_ntoa_r(struct in_addr ina, char *buf); /* in libkern */
char *inet_ntop(int, const void *, char *, socklen_t); /* in libkern */
int inet_pton(int af, const char *, void *); /* in libkern */
void in_ifdetach(struct ifnet *);
#define in_hosteq(s, t) ((s).s_addr == (t).s_addr)
#define in_nullhost(x) ((x).s_addr == INADDR_ANY)
#define in_allhosts(x) ((x).s_addr == htonl(INADDR_ALLHOSTS_GROUP))
#define satosin(sa) ((struct sockaddr_in *)(sa))
#define sintosa(sin) ((struct sockaddr *)(sin))
#define ifatoia(ifa) ((struct in_ifaddr *)(ifa))
#endif /* _KERNEL */
/* INET6 stuff */
#if __POSIX_VISIBLE >= 200112
#define __KAME_NETINET_IN_H_INCLUDED_
#include <netinet6/in6.h>
#undef __KAME_NETINET_IN_H_INCLUDED_
#endif
#endif /* !_NETINET_IN_H_*/
diff --git a/sys/netinet6/in6.c b/sys/netinet6/in6.c
index d54aba58edb6..feee4f3f64a5 100644
--- a/sys/netinet6/in6.c
+++ b/sys/netinet6/in6.c
@@ -1,2584 +1,2605 @@
/*-
* SPDX-License-Identifier: BSD-3-Clause
*
* 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: in6.c,v 1.259 2002/01/21 11:37:50 keiichi Exp $
*/
/*-
* Copyright (c) 1982, 1986, 1991, 1993
* The Regents of the University of California. 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 University 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 REGENTS 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 REGENTS 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.
*
* @(#)in.c 8.2 (Berkeley) 11/15/93
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_inet.h"
#include "opt_inet6.h"
#include <sys/param.h>
#include <sys/eventhandler.h>
#include <sys/errno.h>
#include <sys/jail.h>
#include <sys/malloc.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/sockio.h>
#include <sys/systm.h>
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/protosw.h>
#include <sys/time.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/rmlock.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_types.h>
#include <net/route.h>
#include <net/route/route_ctl.h>
#include <net/route/nhop.h>
#include <net/if_dl.h>
#include <net/vnet.h>
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <net/if_llatbl.h>
#include <netinet/if_ether.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/in_pcb.h>
#include <netinet/ip_carp.h>
#include <netinet/ip6.h>
#include <netinet6/ip6_var.h>
#include <netinet6/nd6.h>
#include <netinet6/mld6_var.h>
#include <netinet6/ip6_mroute.h>
#include <netinet6/in6_ifattach.h>
#include <netinet6/scope6_var.h>
#include <netinet6/in6_fib.h>
#include <netinet6/in6_pcb.h>
/*
* struct in6_ifreq and struct ifreq must be type punnable for common members
* of ifr_ifru to allow accessors to be shared.
*/
_Static_assert(offsetof(struct in6_ifreq, ifr_ifru) ==
offsetof(struct ifreq, ifr_ifru),
"struct in6_ifreq and struct ifreq are not type punnable");
VNET_DECLARE(int, icmp6_nodeinfo_oldmcprefix);
#define V_icmp6_nodeinfo_oldmcprefix VNET(icmp6_nodeinfo_oldmcprefix)
/*
* Definitions of some costant IP6 addresses.
*/
const struct in6_addr in6addr_any = IN6ADDR_ANY_INIT;
const struct in6_addr in6addr_loopback = IN6ADDR_LOOPBACK_INIT;
const struct in6_addr in6addr_nodelocal_allnodes =
IN6ADDR_NODELOCAL_ALLNODES_INIT;
const struct in6_addr in6addr_linklocal_allnodes =
IN6ADDR_LINKLOCAL_ALLNODES_INIT;
const struct in6_addr in6addr_linklocal_allrouters =
IN6ADDR_LINKLOCAL_ALLROUTERS_INIT;
const struct in6_addr in6addr_linklocal_allv2routers =
IN6ADDR_LINKLOCAL_ALLV2ROUTERS_INIT;
const struct in6_addr in6mask0 = IN6MASK0;
const struct in6_addr in6mask32 = IN6MASK32;
const struct in6_addr in6mask64 = IN6MASK64;
const struct in6_addr in6mask96 = IN6MASK96;
const struct in6_addr in6mask128 = IN6MASK128;
const struct sockaddr_in6 sa6_any =
{ sizeof(sa6_any), AF_INET6, 0, 0, IN6ADDR_ANY_INIT, 0 };
static int in6_notify_ifa(struct ifnet *, struct in6_ifaddr *,
struct in6_aliasreq *, int);
static void in6_unlink_ifa(struct in6_ifaddr *, struct ifnet *);
static int in6_validate_ifra(struct ifnet *, struct in6_aliasreq *,
struct in6_ifaddr *, int);
static struct in6_ifaddr *in6_alloc_ifa(struct ifnet *,
struct in6_aliasreq *, int flags);
static int in6_update_ifa_internal(struct ifnet *, struct in6_aliasreq *,
struct in6_ifaddr *, int, int);
static int in6_broadcast_ifa(struct ifnet *, struct in6_aliasreq *,
struct in6_ifaddr *, int);
#define ifa2ia6(ifa) ((struct in6_ifaddr *)(ifa))
#define ia62ifa(ia6) (&((ia6)->ia_ifa))
void
in6_newaddrmsg(struct in6_ifaddr *ia, int cmd)
{
struct rt_addrinfo info;
struct ifaddr *ifa;
struct sockaddr_dl gateway;
int fibnum;
ifa = &ia->ia_ifa;
/*
* Prepare info data for the host route.
* This code mimics one from ifa_maintain_loopback_route().
*/
bzero(&info, sizeof(struct rt_addrinfo));
info.rti_flags = ifa->ifa_flags | RTF_HOST | RTF_STATIC | RTF_PINNED;
info.rti_info[RTAX_DST] = ifa->ifa_addr;
info.rti_info[RTAX_GATEWAY] = (struct sockaddr *)&gateway;
link_init_sdl(ifa->ifa_ifp, (struct sockaddr *)&gateway, ifa->ifa_ifp->if_type);
if (cmd != RTM_DELETE)
info.rti_ifp = V_loif;
fibnum = ia62ifa(ia)->ifa_ifp->if_fib;
if (cmd == RTM_ADD) {
rt_addrmsg(cmd, &ia->ia_ifa, fibnum);
rt_routemsg_info(cmd, &info, fibnum);
} else if (cmd == RTM_DELETE) {
rt_routemsg_info(cmd, &info, fibnum);
rt_addrmsg(cmd, &ia->ia_ifa, fibnum);
}
}
int
in6_mask2len(struct in6_addr *mask, u_char *lim0)
{
int x = 0, y;
u_char *lim = lim0, *p;
/* ignore the scope_id part */
if (lim0 == NULL || lim0 - (u_char *)mask > sizeof(*mask))
lim = (u_char *)mask + sizeof(*mask);
for (p = (u_char *)mask; p < lim; x++, p++) {
if (*p != 0xff)
break;
}
y = 0;
if (p < lim) {
for (y = 0; y < 8; y++) {
if ((*p & (0x80 >> y)) == 0)
break;
}
}
/*
* when the limit pointer is given, do a stricter check on the
* remaining bits.
*/
if (p < lim) {
if (y != 0 && (*p & (0x00ff >> y)) != 0)
return (-1);
for (p = p + 1; p < lim; p++)
if (*p != 0)
return (-1);
}
return x * 8 + y;
}
#ifdef COMPAT_FREEBSD32
struct in6_ndifreq32 {
char ifname[IFNAMSIZ];
uint32_t ifindex;
};
#define SIOCGDEFIFACE32_IN6 _IOWR('i', 86, struct in6_ndifreq32)
#endif
int
in6_control(struct socket *so, u_long cmd, caddr_t data,
struct ifnet *ifp, struct thread *td)
{
struct in6_ifreq *ifr = (struct in6_ifreq *)data;
struct in6_ifaddr *ia = NULL;
struct in6_aliasreq *ifra = (struct in6_aliasreq *)data;
struct sockaddr_in6 *sa6;
int carp_attached = 0;
int error;
u_long ocmd = cmd;
/*
* Compat to make pre-10.x ifconfig(8) operable.
*/
if (cmd == OSIOCAIFADDR_IN6)
cmd = SIOCAIFADDR_IN6;
switch (cmd) {
case SIOCGETSGCNT_IN6:
case SIOCGETMIFCNT_IN6:
/*
* XXX mrt_ioctl has a 3rd, unused, FIB argument in route.c.
* We cannot see how that would be needed, so do not adjust the
* KPI blindly; more likely should clean up the IPv4 variant.
*/
return (mrt6_ioctl ? mrt6_ioctl(cmd, data) : EOPNOTSUPP);
}
switch (cmd) {
case SIOCAADDRCTL_POLICY:
case SIOCDADDRCTL_POLICY:
if (td != NULL) {
error = priv_check(td, PRIV_NETINET_ADDRCTRL6);
if (error)
return (error);
}
return (in6_src_ioctl(cmd, data));
}
if (ifp == NULL)
return (EOPNOTSUPP);
switch (cmd) {
case SIOCSNDFLUSH_IN6:
case SIOCSPFXFLUSH_IN6:
case SIOCSRTRFLUSH_IN6:
case SIOCSDEFIFACE_IN6:
case SIOCSIFINFO_FLAGS:
case SIOCSIFINFO_IN6:
if (td != NULL) {
error = priv_check(td, PRIV_NETINET_ND6);
if (error)
return (error);
}
/* FALLTHROUGH */
case OSIOCGIFINFO_IN6:
case SIOCGIFINFO_IN6:
case SIOCGNBRINFO_IN6:
case SIOCGDEFIFACE_IN6:
return (nd6_ioctl(cmd, data, ifp));
#ifdef COMPAT_FREEBSD32
case SIOCGDEFIFACE32_IN6:
{
struct in6_ndifreq ndif;
struct in6_ndifreq32 *ndif32;
error = nd6_ioctl(SIOCGDEFIFACE_IN6, (caddr_t)&ndif,
ifp);
if (error)
return (error);
ndif32 = (struct in6_ndifreq32 *)data;
ndif32->ifindex = ndif.ifindex;
return (0);
}
#endif
}
switch (cmd) {
case SIOCSIFPREFIX_IN6:
case SIOCDIFPREFIX_IN6:
case SIOCAIFPREFIX_IN6:
case SIOCCIFPREFIX_IN6:
case SIOCSGIFPREFIX_IN6:
case SIOCGIFPREFIX_IN6:
log(LOG_NOTICE,
"prefix ioctls are now invalidated. "
"please use ifconfig.\n");
return (EOPNOTSUPP);
}
switch (cmd) {
case SIOCSSCOPE6:
if (td != NULL) {
error = priv_check(td, PRIV_NETINET_SCOPE6);
if (error)
return (error);
}
/* FALLTHROUGH */
case SIOCGSCOPE6:
case SIOCGSCOPE6DEF:
return (scope6_ioctl(cmd, data, ifp));
}
/*
* Find address for this interface, if it exists.
*
* In netinet code, we have checked ifra_addr in SIOCSIF*ADDR operation
* only, and used the first interface address as the target of other
* operations (without checking ifra_addr). This was because netinet
* code/API assumed at most 1 interface address per interface.
* Since IPv6 allows a node to assign multiple addresses
* on a single interface, we almost always look and check the
* presence of ifra_addr, and reject invalid ones here.
* It also decreases duplicated code among SIOC*_IN6 operations.
*/
switch (cmd) {
case SIOCAIFADDR_IN6:
case SIOCSIFPHYADDR_IN6:
sa6 = &ifra->ifra_addr;
break;
case SIOCSIFADDR_IN6:
case SIOCGIFADDR_IN6:
case SIOCSIFDSTADDR_IN6:
case SIOCSIFNETMASK_IN6:
case SIOCGIFDSTADDR_IN6:
case SIOCGIFNETMASK_IN6:
case SIOCDIFADDR_IN6:
case SIOCGIFPSRCADDR_IN6:
case SIOCGIFPDSTADDR_IN6:
case SIOCGIFAFLAG_IN6:
case SIOCSNDFLUSH_IN6:
case SIOCSPFXFLUSH_IN6:
case SIOCSRTRFLUSH_IN6:
case SIOCGIFALIFETIME_IN6:
case SIOCGIFSTAT_IN6:
case SIOCGIFSTAT_ICMP6:
sa6 = &ifr->ifr_addr;
break;
case SIOCSIFADDR:
case SIOCSIFBRDADDR:
case SIOCSIFDSTADDR:
case SIOCSIFNETMASK:
/*
* Although we should pass any non-INET6 ioctl requests
* down to driver, we filter some legacy INET requests.
* Drivers trust SIOCSIFADDR et al to come from an already
* privileged layer, and do not perform any credentials
* checks or input validation.
*/
return (EINVAL);
default:
sa6 = NULL;
break;
}
if (sa6 && sa6->sin6_family == AF_INET6) {
if (sa6->sin6_scope_id != 0)
error = sa6_embedscope(sa6, 0);
else
error = in6_setscope(&sa6->sin6_addr, ifp, NULL);
if (error != 0)
return (error);
if (td != NULL && (error = prison_check_ip6(td->td_ucred,
&sa6->sin6_addr)) != 0)
return (error);
ia = in6ifa_ifpwithaddr(ifp, &sa6->sin6_addr);
} else
ia = NULL;
switch (cmd) {
case SIOCSIFADDR_IN6:
case SIOCSIFDSTADDR_IN6:
case SIOCSIFNETMASK_IN6:
/*
* Since IPv6 allows a node to assign multiple addresses
* on a single interface, SIOCSIFxxx ioctls are deprecated.
*/
/* we decided to obsolete this command (20000704) */
error = EINVAL;
goto out;
case SIOCDIFADDR_IN6:
/*
* for IPv4, we look for existing in_ifaddr here to allow
* "ifconfig if0 delete" to remove the first IPv4 address on
* the interface. For IPv6, as the spec allows multiple
* interface address from the day one, we consider "remove the
* first one" semantics to be not preferable.
*/
if (ia == NULL) {
error = EADDRNOTAVAIL;
goto out;
}
/* FALLTHROUGH */
case SIOCAIFADDR_IN6:
/*
* We always require users to specify a valid IPv6 address for
* the corresponding operation.
*/
if (ifra->ifra_addr.sin6_family != AF_INET6 ||
ifra->ifra_addr.sin6_len != sizeof(struct sockaddr_in6)) {
error = EAFNOSUPPORT;
goto out;
}
if (td != NULL) {
error = priv_check(td, (cmd == SIOCDIFADDR_IN6) ?
PRIV_NET_DELIFADDR : PRIV_NET_ADDIFADDR);
if (error)
goto out;
}
/* FALLTHROUGH */
case SIOCGIFSTAT_IN6:
case SIOCGIFSTAT_ICMP6:
if (ifp->if_afdata[AF_INET6] == NULL) {
error = EPFNOSUPPORT;
goto out;
}
break;
case SIOCGIFADDR_IN6:
/* This interface is basically deprecated. use SIOCGIFCONF. */
/* FALLTHROUGH */
case SIOCGIFAFLAG_IN6:
case SIOCGIFNETMASK_IN6:
case SIOCGIFDSTADDR_IN6:
case SIOCGIFALIFETIME_IN6:
/* must think again about its semantics */
if (ia == NULL) {
error = EADDRNOTAVAIL;
goto out;
}
break;
}
switch (cmd) {
case SIOCGIFADDR_IN6:
ifr->ifr_addr = ia->ia_addr;
if ((error = sa6_recoverscope(&ifr->ifr_addr)) != 0)
goto out;
break;
case SIOCGIFDSTADDR_IN6:
if ((ifp->if_flags & IFF_POINTOPOINT) == 0) {
error = EINVAL;
goto out;
}
ifr->ifr_dstaddr = ia->ia_dstaddr;
if ((error = sa6_recoverscope(&ifr->ifr_dstaddr)) != 0)
goto out;
break;
case SIOCGIFNETMASK_IN6:
ifr->ifr_addr = ia->ia_prefixmask;
break;
case SIOCGIFAFLAG_IN6:
ifr->ifr_ifru.ifru_flags6 = ia->ia6_flags;
break;
case SIOCGIFSTAT_IN6:
COUNTER_ARRAY_COPY(((struct in6_ifextra *)
ifp->if_afdata[AF_INET6])->in6_ifstat,
&ifr->ifr_ifru.ifru_stat,
sizeof(struct in6_ifstat) / sizeof(uint64_t));
break;
case SIOCGIFSTAT_ICMP6:
COUNTER_ARRAY_COPY(((struct in6_ifextra *)
ifp->if_afdata[AF_INET6])->icmp6_ifstat,
&ifr->ifr_ifru.ifru_icmp6stat,
sizeof(struct icmp6_ifstat) / sizeof(uint64_t));
break;
case SIOCGIFALIFETIME_IN6:
ifr->ifr_ifru.ifru_lifetime = ia->ia6_lifetime;
if (ia->ia6_lifetime.ia6t_vltime != ND6_INFINITE_LIFETIME) {
time_t maxexpire;
struct in6_addrlifetime *retlt =
&ifr->ifr_ifru.ifru_lifetime;
/*
* XXX: adjust expiration time assuming time_t is
* signed.
*/
maxexpire = (-1) &
~((time_t)1 << ((sizeof(maxexpire) * 8) - 1));
if (ia->ia6_lifetime.ia6t_vltime <
maxexpire - ia->ia6_updatetime) {
retlt->ia6t_expire = ia->ia6_updatetime +
ia->ia6_lifetime.ia6t_vltime;
} else
retlt->ia6t_expire = maxexpire;
}
if (ia->ia6_lifetime.ia6t_pltime != ND6_INFINITE_LIFETIME) {
time_t maxexpire;
struct in6_addrlifetime *retlt =
&ifr->ifr_ifru.ifru_lifetime;
/*
* XXX: adjust expiration time assuming time_t is
* signed.
*/
maxexpire = (-1) &
~((time_t)1 << ((sizeof(maxexpire) * 8) - 1));
if (ia->ia6_lifetime.ia6t_pltime <
maxexpire - ia->ia6_updatetime) {
retlt->ia6t_preferred = ia->ia6_updatetime +
ia->ia6_lifetime.ia6t_pltime;
} else
retlt->ia6t_preferred = maxexpire;
}
break;
case SIOCAIFADDR_IN6:
{
struct nd_prefixctl pr0;
struct nd_prefix *pr;
/*
* first, make or update the interface address structure,
* and link it to the list.
*/
if ((error = in6_update_ifa(ifp, ifra, ia, 0)) != 0)
goto out;
if (ia != NULL) {
if (ia->ia_ifa.ifa_carp)
(*carp_detach_p)(&ia->ia_ifa, true);
ifa_free(&ia->ia_ifa);
}
if ((ia = in6ifa_ifpwithaddr(ifp, &ifra->ifra_addr.sin6_addr))
== NULL) {
/*
* this can happen when the user specify the 0 valid
* lifetime.
*/
break;
}
if (cmd == ocmd && ifra->ifra_vhid > 0) {
if (carp_attach_p != NULL)
error = (*carp_attach_p)(&ia->ia_ifa,
ifra->ifra_vhid);
else
error = EPROTONOSUPPORT;
if (error)
goto out;
else
carp_attached = 1;
}
/*
* then, make the prefix on-link on the interface.
* XXX: we'd rather create the prefix before the address, but
* we need at least one address to install the corresponding
* interface route, so we configure the address first.
*/
/*
* convert mask to prefix length (prefixmask has already
* been validated in in6_update_ifa().
*/
bzero(&pr0, sizeof(pr0));
pr0.ndpr_ifp = ifp;
pr0.ndpr_plen = in6_mask2len(&ifra->ifra_prefixmask.sin6_addr,
NULL);
if (pr0.ndpr_plen == 128) {
/* we don't need to install a host route. */
goto aifaddr_out;
}
pr0.ndpr_prefix = ifra->ifra_addr;
/* apply the mask for safety. */
IN6_MASK_ADDR(&pr0.ndpr_prefix.sin6_addr,
&ifra->ifra_prefixmask.sin6_addr);
/*
* XXX: since we don't have an API to set prefix (not address)
* lifetimes, we just use the same lifetimes as addresses.
* The (temporarily) installed lifetimes can be overridden by
* later advertised RAs (when accept_rtadv is non 0), which is
* an intended behavior.
*/
pr0.ndpr_raf_onlink = 1; /* should be configurable? */
pr0.ndpr_raf_auto =
((ifra->ifra_flags & IN6_IFF_AUTOCONF) != 0);
pr0.ndpr_vltime = ifra->ifra_lifetime.ia6t_vltime;
pr0.ndpr_pltime = ifra->ifra_lifetime.ia6t_pltime;
/* add the prefix if not yet. */
if ((pr = nd6_prefix_lookup(&pr0)) == NULL) {
/*
* nd6_prelist_add will install the corresponding
* interface route.
*/
if ((error = nd6_prelist_add(&pr0, NULL, &pr)) != 0) {
if (carp_attached)
(*carp_detach_p)(&ia->ia_ifa, false);
goto out;
}
}
/* relate the address to the prefix */
if (ia->ia6_ndpr == NULL) {
ia->ia6_ndpr = pr;
pr->ndpr_addrcnt++;
/*
* If this is the first autoconf address from the
* prefix, create a temporary address as well
* (when required).
*/
if ((ia->ia6_flags & IN6_IFF_AUTOCONF) &&
V_ip6_use_tempaddr && pr->ndpr_addrcnt == 1) {
int e;
if ((e = in6_tmpifadd(ia, 1, 0)) != 0) {
log(LOG_NOTICE, "in6_control: failed "
"to create a temporary address, "
"errno=%d\n", e);
}
}
}
nd6_prefix_rele(pr);
/*
* this might affect the status of autoconfigured addresses,
* that is, this address might make other addresses detached.
*/
pfxlist_onlink_check();
aifaddr_out:
/*
* Try to clear the flag when a new IPv6 address is added
* onto an IFDISABLED interface and it succeeds.
*/
if (ND_IFINFO(ifp)->flags & ND6_IFF_IFDISABLED) {
struct in6_ndireq nd;
memset(&nd, 0, sizeof(nd));
nd.ndi.flags = ND_IFINFO(ifp)->flags;
nd.ndi.flags &= ~ND6_IFF_IFDISABLED;
if (nd6_ioctl(SIOCSIFINFO_FLAGS, (caddr_t)&nd, ifp) < 0)
log(LOG_NOTICE, "SIOCAIFADDR_IN6: "
"SIOCSIFINFO_FLAGS for -ifdisabled "
"failed.");
/*
* Ignore failure of clearing the flag intentionally.
* The failure means address duplication was detected.
*/
}
break;
}
case SIOCDIFADDR_IN6:
in6_purgeifaddr(ia);
EVENTHANDLER_INVOKE(ifaddr_event_ext, ifp, &ia->ia_ifa,
IFADDR_EVENT_DEL);
break;
default:
if (ifp->if_ioctl == NULL) {
error = EOPNOTSUPP;
goto out;
}
error = (*ifp->if_ioctl)(ifp, cmd, data);
goto out;
}
error = 0;
out:
if (ia != NULL)
ifa_free(&ia->ia_ifa);
return (error);
}
static struct in6_multi_mship *
in6_joingroup_legacy(struct ifnet *ifp, const struct in6_addr *mcaddr,
int *errorp, int delay)
{
struct in6_multi_mship *imm;
int error;
imm = malloc(sizeof(*imm), M_IP6MADDR, M_NOWAIT);
if (imm == NULL) {
*errorp = ENOBUFS;
return (NULL);
}
delay = (delay * PR_FASTHZ) / hz;
error = in6_joingroup(ifp, mcaddr, NULL, &imm->i6mm_maddr, delay);
if (error) {
*errorp = error;
free(imm, M_IP6MADDR);
return (NULL);
}
return (imm);
}
/*
* Join necessary multicast groups. Factored out from in6_update_ifa().
* This entire work should only be done once, for the default FIB.
*/
static int
in6_update_ifa_join_mc(struct ifnet *ifp, struct in6_aliasreq *ifra,
struct in6_ifaddr *ia, int flags, struct in6_multi **in6m_sol)
{
char ip6buf[INET6_ADDRSTRLEN];
struct in6_addr mltaddr;
struct in6_multi_mship *imm;
int delay, error;
KASSERT(in6m_sol != NULL, ("%s: in6m_sol is NULL", __func__));
/* Join solicited multicast addr for new host id. */
bzero(&mltaddr, sizeof(struct in6_addr));
mltaddr.s6_addr32[0] = IPV6_ADDR_INT32_MLL;
mltaddr.s6_addr32[2] = htonl(1);
mltaddr.s6_addr32[3] = ifra->ifra_addr.sin6_addr.s6_addr32[3];
mltaddr.s6_addr8[12] = 0xff;
if ((error = in6_setscope(&mltaddr, ifp, NULL)) != 0) {
/* XXX: should not happen */
log(LOG_ERR, "%s: in6_setscope failed\n", __func__);
goto cleanup;
}
delay = error = 0;
if ((flags & IN6_IFAUPDATE_DADDELAY)) {
/*
* We need a random delay for DAD on the address being
* configured. It also means delaying transmission of the
* corresponding MLD report to avoid report collision.
* [RFC 4861, Section 6.3.7]
*/
delay = arc4random() % (MAX_RTR_SOLICITATION_DELAY * hz);
}
imm = in6_joingroup_legacy(ifp, &mltaddr, &error, delay);
if (imm == NULL) {
nd6log((LOG_WARNING, "%s: in6_joingroup failed for %s on %s "
"(errno=%d)\n", __func__, ip6_sprintf(ip6buf, &mltaddr),
if_name(ifp), error));
goto cleanup;
}
LIST_INSERT_HEAD(&ia->ia6_memberships, imm, i6mm_chain);
*in6m_sol = imm->i6mm_maddr;
/*
* Join link-local all-nodes address.
*/
mltaddr = in6addr_linklocal_allnodes;
if ((error = in6_setscope(&mltaddr, ifp, NULL)) != 0)
goto cleanup; /* XXX: should not fail */
imm = in6_joingroup_legacy(ifp, &mltaddr, &error, 0);
if (imm == NULL) {
nd6log((LOG_WARNING, "%s: in6_joingroup failed for %s on %s "
"(errno=%d)\n", __func__, ip6_sprintf(ip6buf, &mltaddr),
if_name(ifp), error));
goto cleanup;
}
LIST_INSERT_HEAD(&ia->ia6_memberships, imm, i6mm_chain);
/*
* Join node information group address.
*/
delay = 0;
if ((flags & IN6_IFAUPDATE_DADDELAY)) {
/*
* The spec does not say anything about delay for this group,
* but the same logic should apply.
*/
delay = arc4random() % (MAX_RTR_SOLICITATION_DELAY * hz);
}
if (in6_nigroup(ifp, NULL, -1, &mltaddr) == 0) {
/* XXX jinmei */
imm = in6_joingroup_legacy(ifp, &mltaddr, &error, delay);
if (imm == NULL)
nd6log((LOG_WARNING,
"%s: in6_joingroup failed for %s on %s "
"(errno=%d)\n", __func__, ip6_sprintf(ip6buf,
&mltaddr), if_name(ifp), error));
/* XXX not very fatal, go on... */
else
LIST_INSERT_HEAD(&ia->ia6_memberships, imm, i6mm_chain);
}
if (V_icmp6_nodeinfo_oldmcprefix &&
in6_nigroup_oldmcprefix(ifp, NULL, -1, &mltaddr) == 0) {
imm = in6_joingroup_legacy(ifp, &mltaddr, &error, delay);
if (imm == NULL)
nd6log((LOG_WARNING,
"%s: in6_joingroup failed for %s on %s "
"(errno=%d)\n", __func__, ip6_sprintf(ip6buf,
&mltaddr), if_name(ifp), error));
/* XXX not very fatal, go on... */
else
LIST_INSERT_HEAD(&ia->ia6_memberships, imm, i6mm_chain);
}
/*
* Join interface-local all-nodes address.
* (ff01::1%ifN, and ff01::%ifN/32)
*/
mltaddr = in6addr_nodelocal_allnodes;
if ((error = in6_setscope(&mltaddr, ifp, NULL)) != 0)
goto cleanup; /* XXX: should not fail */
imm = in6_joingroup_legacy(ifp, &mltaddr, &error, 0);
if (imm == NULL) {
nd6log((LOG_WARNING, "%s: in6_joingroup failed for %s on %s "
"(errno=%d)\n", __func__, ip6_sprintf(ip6buf,
&mltaddr), if_name(ifp), error));
goto cleanup;
}
LIST_INSERT_HEAD(&ia->ia6_memberships, imm, i6mm_chain);
cleanup:
return (error);
}
/*
* Update parameters of an IPv6 interface address.
* If necessary, a new entry is created and linked into address chains.
* This function is separated from in6_control().
*/
int
in6_update_ifa(struct ifnet *ifp, struct in6_aliasreq *ifra,
struct in6_ifaddr *ia, int flags)
{
int error, hostIsNew = 0;
if ((error = in6_validate_ifra(ifp, ifra, ia, flags)) != 0)
return (error);
if (ia == NULL) {
hostIsNew = 1;
if ((ia = in6_alloc_ifa(ifp, ifra, flags)) == NULL)
return (ENOBUFS);
}
error = in6_update_ifa_internal(ifp, ifra, ia, hostIsNew, flags);
if (error != 0) {
if (hostIsNew != 0) {
in6_unlink_ifa(ia, ifp);
ifa_free(&ia->ia_ifa);
}
return (error);
}
if (hostIsNew)
error = in6_broadcast_ifa(ifp, ifra, ia, flags);
return (error);
}
/*
* Fill in basic IPv6 address request info.
*/
void
in6_prepare_ifra(struct in6_aliasreq *ifra, const struct in6_addr *addr,
const struct in6_addr *mask)
{
memset(ifra, 0, sizeof(struct in6_aliasreq));
ifra->ifra_addr.sin6_family = AF_INET6;
ifra->ifra_addr.sin6_len = sizeof(struct sockaddr_in6);
if (addr != NULL)
ifra->ifra_addr.sin6_addr = *addr;
ifra->ifra_prefixmask.sin6_family = AF_INET6;
ifra->ifra_prefixmask.sin6_len = sizeof(struct sockaddr_in6);
if (mask != NULL)
ifra->ifra_prefixmask.sin6_addr = *mask;
}
static int
in6_validate_ifra(struct ifnet *ifp, struct in6_aliasreq *ifra,
struct in6_ifaddr *ia, int flags)
{
int plen = -1;
struct sockaddr_in6 dst6;
struct in6_addrlifetime *lt;
char ip6buf[INET6_ADDRSTRLEN];
/* Validate parameters */
if (ifp == NULL || ifra == NULL) /* this maybe redundant */
return (EINVAL);
/*
* The destination address for a p2p link must have a family
* of AF_UNSPEC or AF_INET6.
*/
if ((ifp->if_flags & IFF_POINTOPOINT) != 0 &&
ifra->ifra_dstaddr.sin6_family != AF_INET6 &&
ifra->ifra_dstaddr.sin6_family != AF_UNSPEC)
return (EAFNOSUPPORT);
/*
* Validate address
*/
if (ifra->ifra_addr.sin6_len != sizeof(struct sockaddr_in6) ||
ifra->ifra_addr.sin6_family != AF_INET6)
return (EINVAL);
/*
* validate ifra_prefixmask. don't check sin6_family, netmask
* does not carry fields other than sin6_len.
*/
if (ifra->ifra_prefixmask.sin6_len > sizeof(struct sockaddr_in6))
return (EINVAL);
/*
* Because the IPv6 address architecture is classless, we require
* users to specify a (non 0) prefix length (mask) for a new address.
* We also require the prefix (when specified) mask is valid, and thus
* reject a non-consecutive mask.
*/
if (ia == NULL && ifra->ifra_prefixmask.sin6_len == 0)
return (EINVAL);
if (ifra->ifra_prefixmask.sin6_len != 0) {
plen = in6_mask2len(&ifra->ifra_prefixmask.sin6_addr,
(u_char *)&ifra->ifra_prefixmask +
ifra->ifra_prefixmask.sin6_len);
if (plen <= 0)
return (EINVAL);
} else {
/*
* In this case, ia must not be NULL. We just use its prefix
* length.
*/
plen = in6_mask2len(&ia->ia_prefixmask.sin6_addr, NULL);
}
/*
* If the destination address on a p2p interface is specified,
* and the address is a scoped one, validate/set the scope
* zone identifier.
*/
dst6 = ifra->ifra_dstaddr;
if ((ifp->if_flags & (IFF_POINTOPOINT|IFF_LOOPBACK)) != 0 &&
(dst6.sin6_family == AF_INET6)) {
struct in6_addr in6_tmp;
u_int32_t zoneid;
in6_tmp = dst6.sin6_addr;
if (in6_setscope(&in6_tmp, ifp, &zoneid))
return (EINVAL); /* XXX: should be impossible */
if (dst6.sin6_scope_id != 0) {
if (dst6.sin6_scope_id != zoneid)
return (EINVAL);
} else /* user omit to specify the ID. */
dst6.sin6_scope_id = zoneid;
/* convert into the internal form */
if (sa6_embedscope(&dst6, 0))
return (EINVAL); /* XXX: should be impossible */
}
/* Modify original ifra_dstaddr to reflect changes */
ifra->ifra_dstaddr = dst6;
/*
* The destination address can be specified only for a p2p or a
* loopback interface. If specified, the corresponding prefix length
* must be 128.
*/
if (ifra->ifra_dstaddr.sin6_family == AF_INET6) {
if ((ifp->if_flags & (IFF_POINTOPOINT|IFF_LOOPBACK)) == 0) {
/* XXX: noisy message */
nd6log((LOG_INFO, "in6_update_ifa: a destination can "
"be specified for a p2p or a loopback IF only\n"));
return (EINVAL);
}
if (plen != 128) {
nd6log((LOG_INFO, "in6_update_ifa: prefixlen should "
"be 128 when dstaddr is specified\n"));
return (EINVAL);
}
}
/* lifetime consistency check */
lt = &ifra->ifra_lifetime;
if (lt->ia6t_pltime > lt->ia6t_vltime)
return (EINVAL);
if (lt->ia6t_vltime == 0) {
/*
* the following log might be noisy, but this is a typical
* configuration mistake or a tool's bug.
*/
nd6log((LOG_INFO,
"in6_update_ifa: valid lifetime is 0 for %s\n",
ip6_sprintf(ip6buf, &ifra->ifra_addr.sin6_addr)));
if (ia == NULL)
return (0); /* there's nothing to do */
}
/* Check prefix mask */
if (ia != NULL && ifra->ifra_prefixmask.sin6_len != 0) {
/*
* We prohibit changing the prefix length of an existing
* address, because
* + such an operation should be rare in IPv6, and
* + the operation would confuse prefix management.
*/
if (ia->ia_prefixmask.sin6_len != 0 &&
in6_mask2len(&ia->ia_prefixmask.sin6_addr, NULL) != plen) {
nd6log((LOG_INFO, "in6_validate_ifa: the prefix length "
"of an existing %s address should not be changed\n",
ip6_sprintf(ip6buf, &ia->ia_addr.sin6_addr)));
return (EINVAL);
}
}
return (0);
}
/*
* Allocate a new ifaddr and link it into chains.
*/
static struct in6_ifaddr *
in6_alloc_ifa(struct ifnet *ifp, struct in6_aliasreq *ifra, int flags)
{
struct in6_ifaddr *ia;
/*
* When in6_alloc_ifa() is called in a process of a received
* RA, it is called under an interrupt context. So, we should
* call malloc with M_NOWAIT.
*/
ia = (struct in6_ifaddr *)ifa_alloc(sizeof(*ia), M_NOWAIT);
if (ia == NULL)
return (NULL);
LIST_INIT(&ia->ia6_memberships);
/* Initialize the address and masks, and put time stamp */
ia->ia_ifa.ifa_addr = (struct sockaddr *)&ia->ia_addr;
ia->ia_addr.sin6_family = AF_INET6;
ia->ia_addr.sin6_len = sizeof(ia->ia_addr);
/* XXX: Can we assign ,sin6_addr and skip the rest? */
ia->ia_addr = ifra->ifra_addr;
ia->ia6_createtime = time_uptime;
if ((ifp->if_flags & (IFF_POINTOPOINT | IFF_LOOPBACK)) != 0) {
/*
* Some functions expect that ifa_dstaddr is not
* NULL for p2p interfaces.
*/
ia->ia_ifa.ifa_dstaddr =
(struct sockaddr *)&ia->ia_dstaddr;
} else {
ia->ia_ifa.ifa_dstaddr = NULL;
}
/* set prefix mask if any */
ia->ia_ifa.ifa_netmask = (struct sockaddr *)&ia->ia_prefixmask;
if (ifra->ifra_prefixmask.sin6_len != 0) {
ia->ia_prefixmask.sin6_family = AF_INET6;
ia->ia_prefixmask.sin6_len = ifra->ifra_prefixmask.sin6_len;
ia->ia_prefixmask.sin6_addr = ifra->ifra_prefixmask.sin6_addr;
}
ia->ia_ifp = ifp;
ifa_ref(&ia->ia_ifa); /* if_addrhead */
IF_ADDR_WLOCK(ifp);
CK_STAILQ_INSERT_TAIL(&ifp->if_addrhead, &ia->ia_ifa, ifa_link);
IF_ADDR_WUNLOCK(ifp);
ifa_ref(&ia->ia_ifa); /* in6_ifaddrhead */
IN6_IFADDR_WLOCK();
CK_STAILQ_INSERT_TAIL(&V_in6_ifaddrhead, ia, ia_link);
CK_LIST_INSERT_HEAD(IN6ADDR_HASH(&ia->ia_addr.sin6_addr), ia, ia6_hash);
IN6_IFADDR_WUNLOCK();
return (ia);
}
/*
* Update/configure interface address parameters:
*
* 1) Update lifetime
* 2) Update interface metric ad flags
* 3) Notify other subsystems
*/
static int
in6_update_ifa_internal(struct ifnet *ifp, struct in6_aliasreq *ifra,
struct in6_ifaddr *ia, int hostIsNew, int flags)
{
int error;
/* update timestamp */
ia->ia6_updatetime = time_uptime;
/*
* Set lifetimes. We do not refer to ia6t_expire and ia6t_preferred
* to see if the address is deprecated or invalidated, but initialize
* these members for applications.
*/
ia->ia6_lifetime = ifra->ifra_lifetime;
if (ia->ia6_lifetime.ia6t_vltime != ND6_INFINITE_LIFETIME) {
ia->ia6_lifetime.ia6t_expire =
time_uptime + ia->ia6_lifetime.ia6t_vltime;
} else
ia->ia6_lifetime.ia6t_expire = 0;
if (ia->ia6_lifetime.ia6t_pltime != ND6_INFINITE_LIFETIME) {
ia->ia6_lifetime.ia6t_preferred =
time_uptime + ia->ia6_lifetime.ia6t_pltime;
} else
ia->ia6_lifetime.ia6t_preferred = 0;
/*
* backward compatibility - if IN6_IFF_DEPRECATED is set from the
* userland, make it deprecated.
*/
if ((ifra->ifra_flags & IN6_IFF_DEPRECATED) != 0) {
ia->ia6_lifetime.ia6t_pltime = 0;
ia->ia6_lifetime.ia6t_preferred = time_uptime;
}
/*
* configure address flags.
*/
ia->ia6_flags = ifra->ifra_flags;
/*
* Make the address tentative before joining multicast addresses,
* so that corresponding MLD responses would not have a tentative
* source address.
*/
ia->ia6_flags &= ~IN6_IFF_DUPLICATED; /* safety */
/*
* DAD should be performed for an new address or addresses on
* an interface with ND6_IFF_IFDISABLED.
*/
if (in6if_do_dad(ifp) &&
(hostIsNew || (ND_IFINFO(ifp)->flags & ND6_IFF_IFDISABLED)))
ia->ia6_flags |= IN6_IFF_TENTATIVE;
/* notify other subsystems */
error = in6_notify_ifa(ifp, ia, ifra, hostIsNew);
return (error);
}
/*
* Do link-level ifa job:
* 1) Add lle entry for added address
* 2) Notifies routing socket users about new address
* 3) join appropriate multicast group
* 4) start DAD if enabled
*/
static int
in6_broadcast_ifa(struct ifnet *ifp, struct in6_aliasreq *ifra,
struct in6_ifaddr *ia, int flags)
{
struct in6_multi *in6m_sol;
int error = 0;
/* Add local address to lltable, if necessary (ex. on p2p link). */
if ((error = nd6_add_ifa_lle(ia)) != 0) {
in6_purgeaddr(&ia->ia_ifa);
ifa_free(&ia->ia_ifa);
return (error);
}
/* Join necessary multicast groups. */
in6m_sol = NULL;
if ((ifp->if_flags & IFF_MULTICAST) != 0) {
error = in6_update_ifa_join_mc(ifp, ifra, ia, flags, &in6m_sol);
if (error != 0) {
in6_purgeaddr(&ia->ia_ifa);
ifa_free(&ia->ia_ifa);
return (error);
}
}
/* Perform DAD, if the address is TENTATIVE. */
if ((ia->ia6_flags & IN6_IFF_TENTATIVE)) {
int delay, mindelay, maxdelay;
delay = 0;
if ((flags & IN6_IFAUPDATE_DADDELAY)) {
/*
* We need to impose a delay before sending an NS
* for DAD. Check if we also needed a delay for the
* corresponding MLD message. If we did, the delay
* should be larger than the MLD delay (this could be
* relaxed a bit, but this simple logic is at least
* safe).
* XXX: Break data hiding guidelines and look at
* state for the solicited multicast group.
*/
mindelay = 0;
if (in6m_sol != NULL &&
in6m_sol->in6m_state == MLD_REPORTING_MEMBER) {
mindelay = in6m_sol->in6m_timer;
}
maxdelay = MAX_RTR_SOLICITATION_DELAY * hz;
if (maxdelay - mindelay == 0)
delay = 0;
else {
delay =
(arc4random() % (maxdelay - mindelay)) +
mindelay;
}
}
nd6_dad_start((struct ifaddr *)ia, delay);
}
in6_newaddrmsg(ia, RTM_ADD);
ifa_free(&ia->ia_ifa);
return (error);
}
/*
* Adds or deletes interface route for p2p ifa.
* Returns 0 on success or errno.
*/
static int
in6_handle_dstaddr_rtrequest(int cmd, struct in6_ifaddr *ia)
{
struct epoch_tracker et;
struct ifaddr *ifa = &ia->ia_ifa;
int error;
/* Prepare gateway */
struct sockaddr_dl_short sdl = {
.sdl_family = AF_LINK,
.sdl_len = sizeof(struct sockaddr_dl_short),
.sdl_type = ifa->ifa_ifp->if_type,
.sdl_index = ifa->ifa_ifp->if_index,
};
struct sockaddr_in6 dst = {
.sin6_family = AF_INET6,
.sin6_len = sizeof(struct sockaddr_in6),
.sin6_addr = ia->ia_dstaddr.sin6_addr,
};
struct rt_addrinfo info = {
.rti_ifa = ifa,
.rti_flags = RTF_PINNED | RTF_HOST,
.rti_info = {
[RTAX_DST] = (struct sockaddr *)&dst,
[RTAX_GATEWAY] = (struct sockaddr *)&sdl,
},
};
/* Don't set additional per-gw filters on removal */
NET_EPOCH_ENTER(et);
error = rib_handle_ifaddr_info(ifa->ifa_ifp->if_fib, cmd, &info);
NET_EPOCH_EXIT(et);
return (error);
}
static bool
ifa_is_p2p(struct in6_ifaddr *ia)
{
int plen;
plen = in6_mask2len(&ia->ia_prefixmask.sin6_addr, NULL); /* XXX */
if ((plen == 128) && (ia->ia_dstaddr.sin6_family == AF_INET6) &&
!IN6_ARE_ADDR_EQUAL(&ia->ia_addr.sin6_addr, &ia->ia_dstaddr.sin6_addr))
return (true);
return (false);
}
void
in6_purgeaddr(struct ifaddr *ifa)
{
struct ifnet *ifp = ifa->ifa_ifp;
struct in6_ifaddr *ia = (struct in6_ifaddr *) ifa;
struct in6_multi_mship *imm;
int error;
if (ifa->ifa_carp)
(*carp_detach_p)(ifa, false);
/*
* Remove the loopback route to the interface address.
* The check for the current setting of "nd6_useloopback"
* is not needed.
*/
if (ia->ia_flags & IFA_RTSELF) {
error = ifa_del_loopback_route((struct ifaddr *)ia,
(struct sockaddr *)&ia->ia_addr);
if (error == 0)
ia->ia_flags &= ~IFA_RTSELF;
}
/* stop DAD processing */
nd6_dad_stop(ifa);
/* Leave multicast groups. */
while ((imm = LIST_FIRST(&ia->ia6_memberships)) != NULL) {
LIST_REMOVE(imm, i6mm_chain);
if (imm->i6mm_maddr != NULL)
in6_leavegroup(imm->i6mm_maddr, NULL);
free(imm, M_IP6MADDR);
}
/* Check if we need to remove p2p route */
if ((ia->ia_flags & IFA_ROUTE) && ifa_is_p2p(ia)) {
error = in6_handle_dstaddr_rtrequest(RTM_DELETE, ia);
if (error != 0)
log(LOG_INFO, "%s: err=%d, destination address delete "
"failed\n", __func__, error);
ia->ia_flags &= ~IFA_ROUTE;
}
in6_newaddrmsg(ia, RTM_DELETE);
in6_unlink_ifa(ia, ifp);
}
/*
* Removes @ia from the corresponding interfaces and unlinks corresponding
* prefix if no addresses are using it anymore.
*/
void
in6_purgeifaddr(struct in6_ifaddr *ia)
{
struct nd_prefix *pr;
/*
* If the address being deleted is the only one that owns
* the corresponding prefix, expire the prefix as well.
* XXX: theoretically, we don't have to worry about such
* relationship, since we separate the address management
* and the prefix management. We do this, however, to provide
* as much backward compatibility as possible in terms of
* the ioctl operation.
* Note that in6_purgeaddr() will decrement ndpr_addrcnt.
*/
pr = ia->ia6_ndpr;
in6_purgeaddr(&ia->ia_ifa);
if (pr != NULL && pr->ndpr_addrcnt == 0) {
ND6_WLOCK();
nd6_prefix_unlink(pr, NULL);
ND6_WUNLOCK();
nd6_prefix_del(pr);
}
}
static void
in6_unlink_ifa(struct in6_ifaddr *ia, struct ifnet *ifp)
{
char ip6buf[INET6_ADDRSTRLEN];
int remove_lle;
IF_ADDR_WLOCK(ifp);
CK_STAILQ_REMOVE(&ifp->if_addrhead, &ia->ia_ifa, ifaddr, ifa_link);
IF_ADDR_WUNLOCK(ifp);
ifa_free(&ia->ia_ifa); /* if_addrhead */
/*
* Defer the release of what might be the last reference to the
* in6_ifaddr so that it can't be freed before the remainder of the
* cleanup.
*/
IN6_IFADDR_WLOCK();
CK_STAILQ_REMOVE(&V_in6_ifaddrhead, ia, in6_ifaddr, ia_link);
CK_LIST_REMOVE(ia, ia6_hash);
IN6_IFADDR_WUNLOCK();
/*
* Release the reference to the base prefix. There should be a
* positive reference.
*/
remove_lle = 0;
if (ia->ia6_ndpr == NULL) {
nd6log((LOG_NOTICE,
"in6_unlink_ifa: autoconf'ed address "
"%s has no prefix\n", ip6_sprintf(ip6buf, IA6_IN6(ia))));
} else {
ia->ia6_ndpr->ndpr_addrcnt--;
/* Do not delete lles within prefix if refcont != 0 */
if (ia->ia6_ndpr->ndpr_addrcnt == 0)
remove_lle = 1;
ia->ia6_ndpr = NULL;
}
nd6_rem_ifa_lle(ia, remove_lle);
/*
* Also, if the address being removed is autoconf'ed, call
* pfxlist_onlink_check() since the release might affect the status of
* other (detached) addresses.
*/
if ((ia->ia6_flags & IN6_IFF_AUTOCONF)) {
pfxlist_onlink_check();
}
ifa_free(&ia->ia_ifa); /* in6_ifaddrhead */
}
/*
* Notifies other subsystems about address change/arrival:
* 1) Notifies device handler on the first IPv6 address assignment
* 2) Handle routing table changes for P2P links and route
* 3) Handle routing table changes for address host route
*/
static int
in6_notify_ifa(struct ifnet *ifp, struct in6_ifaddr *ia,
struct in6_aliasreq *ifra, int hostIsNew)
{
int error = 0, ifacount = 0;
struct ifaddr *ifa;
struct sockaddr_in6 *pdst;
char ip6buf[INET6_ADDRSTRLEN];
/*
* Give the interface a chance to initialize
* if this is its first address,
*/
if (hostIsNew != 0) {
struct epoch_tracker et;
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
ifacount++;
}
NET_EPOCH_EXIT(et);
}
if (ifacount <= 1 && ifp->if_ioctl) {
error = (*ifp->if_ioctl)(ifp, SIOCSIFADDR, (caddr_t)ia);
if (error)
goto done;
}
/*
* If a new destination address is specified, scrub the old one and
* install the new destination. Note that the interface must be
* p2p or loopback.
*/
pdst = &ifra->ifra_dstaddr;
if (pdst->sin6_family == AF_INET6 &&
!IN6_ARE_ADDR_EQUAL(&pdst->sin6_addr, &ia->ia_dstaddr.sin6_addr)) {
if ((ia->ia_flags & IFA_ROUTE) != 0 &&
(in6_handle_dstaddr_rtrequest(RTM_DELETE, ia) != 0)) {
nd6log((LOG_ERR, "in6_update_ifa_internal: failed to "
"remove a route to the old destination: %s\n",
ip6_sprintf(ip6buf, &ia->ia_addr.sin6_addr)));
/* proceed anyway... */
} else
ia->ia_flags &= ~IFA_ROUTE;
ia->ia_dstaddr = *pdst;
}
/*
* If a new destination address is specified for a point-to-point
* interface, install a route to the destination as an interface
* direct route.
* XXX: the logic below rejects assigning multiple addresses on a p2p
* interface that share the same destination.
*/
if (!(ia->ia_flags & IFA_ROUTE) && ifa_is_p2p(ia)) {
error = in6_handle_dstaddr_rtrequest(RTM_ADD, ia);
if (error)
goto done;
ia->ia_flags |= IFA_ROUTE;
}
/*
* add a loopback route to self if not exists
*/
if (!(ia->ia_flags & IFA_RTSELF) && V_nd6_useloopback) {
error = ifa_add_loopback_route((struct ifaddr *)ia,
(struct sockaddr *)&ia->ia_addr);
if (error == 0)
ia->ia_flags |= IFA_RTSELF;
}
done:
WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, NULL,
"Invoking IPv6 network device address event may sleep");
ifa_ref(&ia->ia_ifa);
EVENTHANDLER_INVOKE(ifaddr_event_ext, ifp, &ia->ia_ifa,
IFADDR_EVENT_ADD);
ifa_free(&ia->ia_ifa);
return (error);
}
/*
* Find an IPv6 interface link-local address specific to an interface.
* ifaddr is returned referenced.
*/
struct in6_ifaddr *
in6ifa_ifpforlinklocal(struct ifnet *ifp, int ignoreflags)
{
struct ifaddr *ifa;
NET_EPOCH_ASSERT();
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
if (IN6_IS_ADDR_LINKLOCAL(IFA_IN6(ifa))) {
if ((((struct in6_ifaddr *)ifa)->ia6_flags &
ignoreflags) != 0)
continue;
ifa_ref(ifa);
break;
}
}
return ((struct in6_ifaddr *)ifa);
}
/*
* find the interface address corresponding to a given IPv6 address.
* ifaddr is returned referenced if @referenced flag is set.
*/
struct in6_ifaddr *
in6ifa_ifwithaddr(const struct in6_addr *addr, uint32_t zoneid, bool referenced)
{
struct rm_priotracker in6_ifa_tracker;
struct in6_ifaddr *ia;
IN6_IFADDR_RLOCK(&in6_ifa_tracker);
CK_LIST_FOREACH(ia, IN6ADDR_HASH(addr), ia6_hash) {
if (IN6_ARE_ADDR_EQUAL(IA6_IN6(ia), addr)) {
if (zoneid != 0 &&
zoneid != ia->ia_addr.sin6_scope_id)
continue;
if (referenced)
ifa_ref(&ia->ia_ifa);
break;
}
}
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return (ia);
}
/*
* find the internet address corresponding to a given interface and address.
* ifaddr is returned referenced.
*/
struct in6_ifaddr *
in6ifa_ifpwithaddr(struct ifnet *ifp, const struct in6_addr *addr)
{
struct epoch_tracker et;
struct ifaddr *ifa;
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
if (IN6_ARE_ADDR_EQUAL(addr, IFA_IN6(ifa))) {
ifa_ref(ifa);
break;
}
}
NET_EPOCH_EXIT(et);
return ((struct in6_ifaddr *)ifa);
}
/*
* Find a link-local scoped address on ifp and return it if any.
*/
struct in6_ifaddr *
in6ifa_llaonifp(struct ifnet *ifp)
{
struct epoch_tracker et;
struct sockaddr_in6 *sin6;
struct ifaddr *ifa;
if (ND_IFINFO(ifp)->flags & ND6_IFF_IFDISABLED)
return (NULL);
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
sin6 = (struct sockaddr_in6 *)ifa->ifa_addr;
if (IN6_IS_SCOPE_LINKLOCAL(&sin6->sin6_addr) ||
IN6_IS_ADDR_MC_INTFACELOCAL(&sin6->sin6_addr) ||
IN6_IS_ADDR_MC_NODELOCAL(&sin6->sin6_addr))
break;
}
NET_EPOCH_EXIT(et);
return ((struct in6_ifaddr *)ifa);
}
/*
* Convert IP6 address to printable (loggable) representation. Caller
* has to make sure that ip6buf is at least INET6_ADDRSTRLEN long.
*/
static char digits[] = "0123456789abcdef";
char *
ip6_sprintf(char *ip6buf, const struct in6_addr *addr)
{
int i, cnt = 0, maxcnt = 0, idx = 0, index = 0;
char *cp;
const u_int16_t *a = (const u_int16_t *)addr;
const u_int8_t *d;
int dcolon = 0, zero = 0;
cp = ip6buf;
for (i = 0; i < 8; i++) {
if (*(a + i) == 0) {
cnt++;
if (cnt == 1)
idx = i;
}
else if (maxcnt < cnt) {
maxcnt = cnt;
index = idx;
cnt = 0;
}
}
if (maxcnt < cnt) {
maxcnt = cnt;
index = idx;
}
for (i = 0; i < 8; i++) {
if (dcolon == 1) {
if (*a == 0) {
if (i == 7)
*cp++ = ':';
a++;
continue;
} else
dcolon = 2;
}
if (*a == 0) {
if (dcolon == 0 && *(a + 1) == 0 && i == index) {
if (i == 0)
*cp++ = ':';
*cp++ = ':';
dcolon = 1;
} else {
*cp++ = '0';
*cp++ = ':';
}
a++;
continue;
}
d = (const u_char *)a;
/* Try to eliminate leading zeros in printout like in :0001. */
zero = 1;
*cp = digits[*d >> 4];
if (*cp != '0') {
zero = 0;
cp++;
}
*cp = digits[*d++ & 0xf];
if (zero == 0 || (*cp != '0')) {
zero = 0;
cp++;
}
*cp = digits[*d >> 4];
if (zero == 0 || (*cp != '0')) {
zero = 0;
cp++;
}
*cp++ = digits[*d & 0xf];
*cp++ = ':';
a++;
}
*--cp = '\0';
return (ip6buf);
}
int
in6_localaddr(struct in6_addr *in6)
{
struct rm_priotracker in6_ifa_tracker;
struct in6_ifaddr *ia;
if (IN6_IS_ADDR_LOOPBACK(in6) || IN6_IS_ADDR_LINKLOCAL(in6))
return 1;
IN6_IFADDR_RLOCK(&in6_ifa_tracker);
CK_STAILQ_FOREACH(ia, &V_in6_ifaddrhead, ia_link) {
if (IN6_ARE_MASKED_ADDR_EQUAL(in6, &ia->ia_addr.sin6_addr,
&ia->ia_prefixmask.sin6_addr)) {
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return 1;
}
}
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return (0);
}
/*
* Return 1 if an internet address is for the local host and configured
* on one of its interfaces.
*/
int
in6_localip(struct in6_addr *in6)
{
struct rm_priotracker in6_ifa_tracker;
struct in6_ifaddr *ia;
IN6_IFADDR_RLOCK(&in6_ifa_tracker);
CK_LIST_FOREACH(ia, IN6ADDR_HASH(in6), ia6_hash) {
if (IN6_ARE_ADDR_EQUAL(in6, &ia->ia_addr.sin6_addr)) {
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return (1);
}
}
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return (0);
}
+/*
+ * Like in6_localip(), but FIB-aware.
+ */
+bool
+in6_localip_fib(struct in6_addr *in6, uint16_t fib)
+{
+ struct rm_priotracker in6_ifa_tracker;
+ struct in6_ifaddr *ia;
+
+ IN6_IFADDR_RLOCK(&in6_ifa_tracker);
+ CK_LIST_FOREACH(ia, IN6ADDR_HASH(in6), ia6_hash) {
+ if (IN6_ARE_ADDR_EQUAL(in6, &ia->ia_addr.sin6_addr) &&
+ ia->ia_ifa.ifa_ifp->if_fib == fib) {
+ IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
+ return (true);
+ }
+ }
+ IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
+ return (false);
+}
+
/*
* Return 1 if an internet address is configured on an interface.
*/
int
in6_ifhasaddr(struct ifnet *ifp, struct in6_addr *addr)
{
struct in6_addr in6;
struct ifaddr *ifa;
struct in6_ifaddr *ia6;
NET_EPOCH_ASSERT();
in6 = *addr;
if (in6_clearscope(&in6))
return (0);
in6_setscope(&in6, ifp, NULL);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
ia6 = (struct in6_ifaddr *)ifa;
if (IN6_ARE_ADDR_EQUAL(&ia6->ia_addr.sin6_addr, &in6))
return (1);
}
return (0);
}
int
in6_is_addr_deprecated(struct sockaddr_in6 *sa6)
{
struct rm_priotracker in6_ifa_tracker;
struct in6_ifaddr *ia;
IN6_IFADDR_RLOCK(&in6_ifa_tracker);
CK_LIST_FOREACH(ia, IN6ADDR_HASH(&sa6->sin6_addr), ia6_hash) {
if (IN6_ARE_ADDR_EQUAL(IA6_IN6(ia), &sa6->sin6_addr)) {
if (ia->ia6_flags & IN6_IFF_DEPRECATED) {
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return (1); /* true */
}
break;
}
}
IN6_IFADDR_RUNLOCK(&in6_ifa_tracker);
return (0); /* false */
}
/*
* return length of part which dst and src are equal
* hard coding...
*/
int
in6_matchlen(struct in6_addr *src, struct in6_addr *dst)
{
int match = 0;
u_char *s = (u_char *)src, *d = (u_char *)dst;
u_char *lim = s + 16, r;
while (s < lim)
if ((r = (*d++ ^ *s++)) != 0) {
while (r < 128) {
match++;
r <<= 1;
}
break;
} else
match += 8;
return match;
}
/* XXX: to be scope conscious */
int
in6_are_prefix_equal(struct in6_addr *p1, struct in6_addr *p2, int len)
{
int bytelen, bitlen;
/* sanity check */
if (0 > len || len > 128) {
log(LOG_ERR, "in6_are_prefix_equal: invalid prefix length(%d)\n",
len);
return (0);
}
bytelen = len / 8;
bitlen = len % 8;
if (bcmp(&p1->s6_addr, &p2->s6_addr, bytelen))
return (0);
if (bitlen != 0 &&
p1->s6_addr[bytelen] >> (8 - bitlen) !=
p2->s6_addr[bytelen] >> (8 - bitlen))
return (0);
return (1);
}
void
in6_prefixlen2mask(struct in6_addr *maskp, int len)
{
u_char maskarray[8] = {0x80, 0xc0, 0xe0, 0xf0, 0xf8, 0xfc, 0xfe, 0xff};
int bytelen, bitlen, i;
/* sanity check */
if (0 > len || len > 128) {
log(LOG_ERR, "in6_prefixlen2mask: invalid prefix length(%d)\n",
len);
return;
}
bzero(maskp, sizeof(*maskp));
bytelen = len / 8;
bitlen = len % 8;
for (i = 0; i < bytelen; i++)
maskp->s6_addr[i] = 0xff;
if (bitlen)
maskp->s6_addr[bytelen] = maskarray[bitlen - 1];
}
/*
* return the best address out of the same scope. if no address was
* found, return the first valid address from designated IF.
*/
struct in6_ifaddr *
in6_ifawithifp(struct ifnet *ifp, struct in6_addr *dst)
{
int dst_scope = in6_addrscope(dst), blen = -1, tlen;
struct ifaddr *ifa;
struct in6_ifaddr *besta = NULL;
struct in6_ifaddr *dep[2]; /* last-resort: deprecated */
NET_EPOCH_ASSERT();
dep[0] = dep[1] = NULL;
/*
* We first look for addresses in the same scope.
* If there is one, return it.
* If two or more, return one which matches the dst longest.
* If none, return one of global addresses assigned other ifs.
*/
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_ANYCAST)
continue; /* XXX: is there any case to allow anycast? */
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_NOTREADY)
continue; /* don't use this interface */
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_DETACHED)
continue;
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_DEPRECATED) {
if (V_ip6_use_deprecated)
dep[0] = (struct in6_ifaddr *)ifa;
continue;
}
if (dst_scope == in6_addrscope(IFA_IN6(ifa))) {
/*
* call in6_matchlen() as few as possible
*/
if (besta) {
if (blen == -1)
blen = in6_matchlen(&besta->ia_addr.sin6_addr, dst);
tlen = in6_matchlen(IFA_IN6(ifa), dst);
if (tlen > blen) {
blen = tlen;
besta = (struct in6_ifaddr *)ifa;
}
} else
besta = (struct in6_ifaddr *)ifa;
}
}
if (besta)
return (besta);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_ANYCAST)
continue; /* XXX: is there any case to allow anycast? */
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_NOTREADY)
continue; /* don't use this interface */
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_DETACHED)
continue;
if (((struct in6_ifaddr *)ifa)->ia6_flags & IN6_IFF_DEPRECATED) {
if (V_ip6_use_deprecated)
dep[1] = (struct in6_ifaddr *)ifa;
continue;
}
return (struct in6_ifaddr *)ifa;
}
/* use the last-resort values, that are, deprecated addresses */
if (dep[0])
return dep[0];
if (dep[1])
return dep[1];
return NULL;
}
/*
* perform DAD when interface becomes IFF_UP.
*/
void
in6_if_up(struct ifnet *ifp)
{
struct epoch_tracker et;
struct ifaddr *ifa;
struct in6_ifaddr *ia;
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
if (ifa->ifa_addr->sa_family != AF_INET6)
continue;
ia = (struct in6_ifaddr *)ifa;
if (ia->ia6_flags & IN6_IFF_TENTATIVE) {
/*
* The TENTATIVE flag was likely set by hand
* beforehand, implicitly indicating the need for DAD.
* We may be able to skip the random delay in this
* case, but we impose delays just in case.
*/
nd6_dad_start(ifa,
arc4random() % (MAX_RTR_SOLICITATION_DELAY * hz));
}
}
NET_EPOCH_EXIT(et);
/*
* special cases, like 6to4, are handled in in6_ifattach
*/
in6_ifattach(ifp, NULL);
}
int
in6if_do_dad(struct ifnet *ifp)
{
if ((ifp->if_flags & IFF_LOOPBACK) != 0)
return (0);
if ((ifp->if_flags & IFF_MULTICAST) == 0)
return (0);
if ((ND_IFINFO(ifp)->flags &
(ND6_IFF_IFDISABLED | ND6_IFF_NO_DAD)) != 0)
return (0);
return (1);
}
/*
* Calculate max IPv6 MTU through all the interfaces and store it
* to in6_maxmtu.
*/
void
in6_setmaxmtu(void)
{
struct epoch_tracker et;
unsigned long maxmtu = 0;
struct ifnet *ifp;
NET_EPOCH_ENTER(et);
CK_STAILQ_FOREACH(ifp, &V_ifnet, if_link) {
/* this function can be called during ifnet initialization */
if (!ifp->if_afdata[AF_INET6])
continue;
if ((ifp->if_flags & IFF_LOOPBACK) == 0 &&
IN6_LINKMTU(ifp) > maxmtu)
maxmtu = IN6_LINKMTU(ifp);
}
NET_EPOCH_EXIT(et);
if (maxmtu) /* update only when maxmtu is positive */
V_in6_maxmtu = maxmtu;
}
/*
* Provide the length of interface identifiers to be used for the link attached
* to the given interface. The length should be defined in "IPv6 over
* xxx-link" document. Note that address architecture might also define
* the length for a particular set of address prefixes, regardless of the
* link type. As clarified in rfc2462bis, those two definitions should be
* consistent, and those really are as of August 2004.
*/
int
in6_if2idlen(struct ifnet *ifp)
{
switch (ifp->if_type) {
case IFT_ETHER: /* RFC2464 */
case IFT_PROPVIRTUAL: /* XXX: no RFC. treat it as ether */
case IFT_L2VLAN: /* ditto */
case IFT_BRIDGE: /* bridge(4) only does Ethernet-like links */
case IFT_INFINIBAND:
return (64);
case IFT_PPP: /* RFC2472 */
return (64);
case IFT_FRELAY: /* RFC2590 */
return (64);
case IFT_IEEE1394: /* RFC3146 */
return (64);
case IFT_GIF:
return (64); /* draft-ietf-v6ops-mech-v2-07 */
case IFT_LOOP:
return (64); /* XXX: is this really correct? */
default:
/*
* Unknown link type:
* It might be controversial to use the today's common constant
* of 64 for these cases unconditionally. For full compliance,
* we should return an error in this case. On the other hand,
* if we simply miss the standard for the link type or a new
* standard is defined for a new link type, the IFID length
* is very likely to be the common constant. As a compromise,
* we always use the constant, but make an explicit notice
* indicating the "unknown" case.
*/
printf("in6_if2idlen: unknown link type (%d)\n", ifp->if_type);
return (64);
}
}
struct in6_llentry {
struct llentry base;
};
#define IN6_LLTBL_DEFAULT_HSIZE 32
#define IN6_LLTBL_HASH(k, h) \
(((((((k >> 8) ^ k) >> 8) ^ k) >> 8) ^ k) & ((h) - 1))
/*
* Do actual deallocation of @lle.
*/
static void
in6_lltable_destroy_lle_unlocked(epoch_context_t ctx)
{
struct llentry *lle;
lle = __containerof(ctx, struct llentry, lle_epoch_ctx);
LLE_LOCK_DESTROY(lle);
LLE_REQ_DESTROY(lle);
free(lle, M_LLTABLE);
}
/*
* Called by LLE_FREE_LOCKED when number of references
* drops to zero.
*/
static void
in6_lltable_destroy_lle(struct llentry *lle)
{
LLE_WUNLOCK(lle);
NET_EPOCH_CALL(in6_lltable_destroy_lle_unlocked, &lle->lle_epoch_ctx);
}
static struct llentry *
in6_lltable_new(const struct in6_addr *addr6, u_int flags)
{
struct in6_llentry *lle;
lle = malloc(sizeof(struct in6_llentry), M_LLTABLE, M_NOWAIT | M_ZERO);
if (lle == NULL) /* NB: caller generates msg */
return NULL;
lle->base.r_l3addr.addr6 = *addr6;
lle->base.lle_refcnt = 1;
lle->base.lle_free = in6_lltable_destroy_lle;
LLE_LOCK_INIT(&lle->base);
LLE_REQ_INIT(&lle->base);
callout_init(&lle->base.lle_timer, 1);
return (&lle->base);
}
static int
in6_lltable_match_prefix(const struct sockaddr *saddr,
const struct sockaddr *smask, u_int flags, struct llentry *lle)
{
const struct in6_addr *addr, *mask, *lle_addr;
addr = &((const struct sockaddr_in6 *)saddr)->sin6_addr;
mask = &((const struct sockaddr_in6 *)smask)->sin6_addr;
lle_addr = &lle->r_l3addr.addr6;
if (IN6_ARE_MASKED_ADDR_EQUAL(lle_addr, addr, mask) == 0)
return (0);
if (lle->la_flags & LLE_IFADDR) {
/*
* Delete LLE_IFADDR records IFF address & flag matches.
* Note that addr is the interface address within prefix
* being matched.
*/
if (IN6_ARE_ADDR_EQUAL(addr, lle_addr) &&
(flags & LLE_STATIC) != 0)
return (1);
return (0);
}
/* flags & LLE_STATIC means deleting both dynamic and static entries */
if ((flags & LLE_STATIC) || !(lle->la_flags & LLE_STATIC))
return (1);
return (0);
}
static void
in6_lltable_free_entry(struct lltable *llt, struct llentry *lle)
{
struct ifnet *ifp;
LLE_WLOCK_ASSERT(lle);
KASSERT(llt != NULL, ("lltable is NULL"));
/* Unlink entry from table */
if ((lle->la_flags & LLE_LINKED) != 0) {
ifp = llt->llt_ifp;
IF_AFDATA_WLOCK_ASSERT(ifp);
lltable_unlink_entry(llt, lle);
}
llentry_free(lle);
}
static int
in6_lltable_rtcheck(struct ifnet *ifp,
u_int flags,
const struct sockaddr *l3addr)
{
const struct sockaddr_in6 *sin6;
struct nhop_object *nh;
struct in6_addr dst;
uint32_t scopeid;
char ip6buf[INET6_ADDRSTRLEN];
int fibnum;
NET_EPOCH_ASSERT();
KASSERT(l3addr->sa_family == AF_INET6,
("sin_family %d", l3addr->sa_family));
sin6 = (const struct sockaddr_in6 *)l3addr;
in6_splitscope(&sin6->sin6_addr, &dst, &scopeid);
fibnum = V_rt_add_addr_allfibs ? RT_DEFAULT_FIB : ifp->if_fib;
nh = fib6_lookup(fibnum, &dst, scopeid, NHR_NONE, 0);
if (nh && ((nh->nh_flags & NHF_GATEWAY) || nh->nh_ifp != ifp)) {
struct ifaddr *ifa;
/*
* Create an ND6 cache for an IPv6 neighbor
* that is not covered by our own prefix.
*/
ifa = ifaof_ifpforaddr(l3addr, ifp);
if (ifa != NULL) {
return 0;
}
log(LOG_INFO, "IPv6 address: \"%s\" is not on the network\n",
ip6_sprintf(ip6buf, &sin6->sin6_addr));
return EINVAL;
}
return 0;
}
static inline uint32_t
in6_lltable_hash_dst(const struct in6_addr *dst, uint32_t hsize)
{
return (IN6_LLTBL_HASH(dst->s6_addr32[3], hsize));
}
static uint32_t
in6_lltable_hash(const struct llentry *lle, uint32_t hsize)
{
return (in6_lltable_hash_dst(&lle->r_l3addr.addr6, hsize));
}
static void
in6_lltable_fill_sa_entry(const struct llentry *lle, struct sockaddr *sa)
{
struct sockaddr_in6 *sin6;
sin6 = (struct sockaddr_in6 *)sa;
bzero(sin6, sizeof(*sin6));
sin6->sin6_family = AF_INET6;
sin6->sin6_len = sizeof(*sin6);
sin6->sin6_addr = lle->r_l3addr.addr6;
}
static inline struct llentry *
in6_lltable_find_dst(struct lltable *llt, const struct in6_addr *dst)
{
struct llentry *lle;
struct llentries *lleh;
u_int hashidx;
hashidx = in6_lltable_hash_dst(dst, llt->llt_hsize);
lleh = &llt->lle_head[hashidx];
CK_LIST_FOREACH(lle, lleh, lle_next) {
if (lle->la_flags & LLE_DELETED)
continue;
if (IN6_ARE_ADDR_EQUAL(&lle->r_l3addr.addr6, dst))
break;
}
return (lle);
}
static void
in6_lltable_delete_entry(struct lltable *llt, struct llentry *lle)
{
lle->la_flags |= LLE_DELETED;
EVENTHANDLER_INVOKE(lle_event, lle, LLENTRY_DELETED);
#ifdef DIAGNOSTIC
log(LOG_INFO, "ifaddr cache = %p is deleted\n", lle);
#endif
llentry_free(lle);
}
static struct llentry *
in6_lltable_alloc(struct lltable *llt, u_int flags,
const struct sockaddr *l3addr)
{
const struct sockaddr_in6 *sin6 = (const struct sockaddr_in6 *)l3addr;
struct ifnet *ifp = llt->llt_ifp;
struct llentry *lle;
char linkhdr[LLE_MAX_LINKHDR];
size_t linkhdrsize;
int lladdr_off;
KASSERT(l3addr->sa_family == AF_INET6,
("sin_family %d", l3addr->sa_family));
/*
* A route that covers the given address must have
* been installed 1st because we are doing a resolution,
* verify this.
*/
if (!(flags & LLE_IFADDR) &&
in6_lltable_rtcheck(ifp, flags, l3addr) != 0)
return (NULL);
lle = in6_lltable_new(&sin6->sin6_addr, flags);
if (lle == NULL) {
log(LOG_INFO, "lla_lookup: new lle malloc failed\n");
return (NULL);
}
lle->la_flags = flags;
if ((flags & LLE_IFADDR) == LLE_IFADDR) {
linkhdrsize = LLE_MAX_LINKHDR;
if (lltable_calc_llheader(ifp, AF_INET6, IF_LLADDR(ifp),
linkhdr, &linkhdrsize, &lladdr_off) != 0) {
NET_EPOCH_CALL(in6_lltable_destroy_lle_unlocked, &lle->lle_epoch_ctx);
return (NULL);
}
lltable_set_entry_addr(ifp, lle, linkhdr, linkhdrsize,
lladdr_off);
lle->la_flags |= LLE_STATIC;
}
if ((lle->la_flags & LLE_STATIC) != 0)
lle->ln_state = ND6_LLINFO_REACHABLE;
return (lle);
}
static struct llentry *
in6_lltable_lookup(struct lltable *llt, u_int flags,
const struct sockaddr *l3addr)
{
const struct sockaddr_in6 *sin6 = (const struct sockaddr_in6 *)l3addr;
int family = flags >> 16;
struct llentry *lle;
IF_AFDATA_LOCK_ASSERT(llt->llt_ifp);
KASSERT(l3addr->sa_family == AF_INET6,
("sin_family %d", l3addr->sa_family));
KASSERT((flags & (LLE_UNLOCKED | LLE_EXCLUSIVE)) !=
(LLE_UNLOCKED | LLE_EXCLUSIVE),
("wrong lle request flags: %#x", flags));
lle = in6_lltable_find_dst(llt, &sin6->sin6_addr);
if (__predict_false(family != AF_INET6))
lle = llentry_lookup_family(lle, family);
if (lle == NULL)
return (NULL);
if (flags & LLE_UNLOCKED)
return (lle);
if (flags & LLE_EXCLUSIVE)
LLE_WLOCK(lle);
else
LLE_RLOCK(lle);
/*
* If the afdata lock is not held, the LLE may have been unlinked while
* we were blocked on the LLE lock. Check for this case.
*/
if (__predict_false((lle->la_flags & LLE_LINKED) == 0)) {
if (flags & LLE_EXCLUSIVE)
LLE_WUNLOCK(lle);
else
LLE_RUNLOCK(lle);
return (NULL);
}
return (lle);
}
static int
in6_lltable_dump_entry(struct lltable *llt, struct llentry *lle,
struct sysctl_req *wr)
{
struct ifnet *ifp = llt->llt_ifp;
/* XXX stack use */
struct {
struct rt_msghdr rtm;
struct sockaddr_in6 sin6;
/*
* ndp.c assumes that sdl is word aligned
*/
#ifdef __LP64__
uint32_t pad;
#endif
struct sockaddr_dl sdl;
} ndpc;
struct sockaddr_dl *sdl;
int error;
bzero(&ndpc, sizeof(ndpc));
/* skip deleted entries */
if ((lle->la_flags & LLE_DELETED) == LLE_DELETED)
return (0);
/* Skip if jailed and not a valid IP of the prison. */
lltable_fill_sa_entry(lle, (struct sockaddr *)&ndpc.sin6);
if (prison_if(wr->td->td_ucred, (struct sockaddr *)&ndpc.sin6) != 0)
return (0);
/*
* produce a msg made of:
* struct rt_msghdr;
* struct sockaddr_in6 (IPv6)
* struct sockaddr_dl;
*/
ndpc.rtm.rtm_msglen = sizeof(ndpc);
ndpc.rtm.rtm_version = RTM_VERSION;
ndpc.rtm.rtm_type = RTM_GET;
ndpc.rtm.rtm_flags = RTF_UP;
ndpc.rtm.rtm_addrs = RTA_DST | RTA_GATEWAY;
sa6_recoverscope(&ndpc.sin6);
/* publish */
if (lle->la_flags & LLE_PUB)
ndpc.rtm.rtm_flags |= RTF_ANNOUNCE;
sdl = &ndpc.sdl;
sdl->sdl_family = AF_LINK;
sdl->sdl_len = sizeof(*sdl);
sdl->sdl_index = ifp->if_index;
sdl->sdl_type = ifp->if_type;
if ((lle->la_flags & LLE_VALID) == LLE_VALID) {
sdl->sdl_alen = ifp->if_addrlen;
bcopy(lle->ll_addr, LLADDR(sdl), ifp->if_addrlen);
} else {
sdl->sdl_alen = 0;
bzero(LLADDR(sdl), ifp->if_addrlen);
}
if (lle->la_expire != 0)
ndpc.rtm.rtm_rmx.rmx_expire = lle->la_expire +
lle->lle_remtime / hz + time_second - time_uptime;
ndpc.rtm.rtm_flags |= (RTF_HOST | RTF_LLDATA);
if (lle->la_flags & LLE_STATIC)
ndpc.rtm.rtm_flags |= RTF_STATIC;
if (lle->la_flags & LLE_IFADDR)
ndpc.rtm.rtm_flags |= RTF_PINNED;
if (lle->ln_router != 0)
ndpc.rtm.rtm_flags |= RTF_GATEWAY;
ndpc.rtm.rtm_rmx.rmx_pksent = lle->la_asked;
/* Store state in rmx_weight value */
ndpc.rtm.rtm_rmx.rmx_state = lle->ln_state;
ndpc.rtm.rtm_index = ifp->if_index;
error = SYSCTL_OUT(wr, &ndpc, sizeof(ndpc));
return (error);
}
static struct lltable *
in6_lltattach(struct ifnet *ifp)
{
struct lltable *llt;
llt = lltable_allocate_htbl(IN6_LLTBL_DEFAULT_HSIZE);
llt->llt_af = AF_INET6;
llt->llt_ifp = ifp;
llt->llt_lookup = in6_lltable_lookup;
llt->llt_alloc_entry = in6_lltable_alloc;
llt->llt_delete_entry = in6_lltable_delete_entry;
llt->llt_dump_entry = in6_lltable_dump_entry;
llt->llt_hash = in6_lltable_hash;
llt->llt_fill_sa_entry = in6_lltable_fill_sa_entry;
llt->llt_free_entry = in6_lltable_free_entry;
llt->llt_match_prefix = in6_lltable_match_prefix;
llt->llt_mark_used = llentry_mark_used;
lltable_link(llt);
return (llt);
}
void *
in6_domifattach(struct ifnet *ifp)
{
struct in6_ifextra *ext;
/* There are not IPv6-capable interfaces. */
switch (ifp->if_type) {
case IFT_PFLOG:
case IFT_PFSYNC:
case IFT_USB:
return (NULL);
}
ext = (struct in6_ifextra *)malloc(sizeof(*ext), M_IFADDR, M_WAITOK);
bzero(ext, sizeof(*ext));
ext->in6_ifstat = malloc(sizeof(counter_u64_t) *
sizeof(struct in6_ifstat) / sizeof(uint64_t), M_IFADDR, M_WAITOK);
COUNTER_ARRAY_ALLOC(ext->in6_ifstat,
sizeof(struct in6_ifstat) / sizeof(uint64_t), M_WAITOK);
ext->icmp6_ifstat = malloc(sizeof(counter_u64_t) *
sizeof(struct icmp6_ifstat) / sizeof(uint64_t), M_IFADDR,
M_WAITOK);
COUNTER_ARRAY_ALLOC(ext->icmp6_ifstat,
sizeof(struct icmp6_ifstat) / sizeof(uint64_t), M_WAITOK);
ext->nd_ifinfo = nd6_ifattach(ifp);
ext->scope6_id = scope6_ifattach(ifp);
ext->lltable = in6_lltattach(ifp);
ext->mld_ifinfo = mld_domifattach(ifp);
return ext;
}
int
in6_domifmtu(struct ifnet *ifp)
{
if (ifp->if_afdata[AF_INET6] == NULL)
return ifp->if_mtu;
return (IN6_LINKMTU(ifp));
}
void
in6_domifdetach(struct ifnet *ifp, void *aux)
{
struct in6_ifextra *ext = (struct in6_ifextra *)aux;
mld_domifdetach(ifp);
scope6_ifdetach(ext->scope6_id);
nd6_ifdetach(ifp, ext->nd_ifinfo);
lltable_free(ext->lltable);
COUNTER_ARRAY_FREE(ext->in6_ifstat,
sizeof(struct in6_ifstat) / sizeof(uint64_t));
free(ext->in6_ifstat, M_IFADDR);
COUNTER_ARRAY_FREE(ext->icmp6_ifstat,
sizeof(struct icmp6_ifstat) / sizeof(uint64_t));
free(ext->icmp6_ifstat, M_IFADDR);
free(ext, M_IFADDR);
}
/*
* Convert sockaddr_in6 to sockaddr_in. Original sockaddr_in6 must be
* v4 mapped addr or v4 compat addr
*/
void
in6_sin6_2_sin(struct sockaddr_in *sin, struct sockaddr_in6 *sin6)
{
bzero(sin, sizeof(*sin));
sin->sin_len = sizeof(struct sockaddr_in);
sin->sin_family = AF_INET;
sin->sin_port = sin6->sin6_port;
sin->sin_addr.s_addr = sin6->sin6_addr.s6_addr32[3];
}
/* Convert sockaddr_in to sockaddr_in6 in v4 mapped addr format. */
void
in6_sin_2_v4mapsin6(struct sockaddr_in *sin, struct sockaddr_in6 *sin6)
{
bzero(sin6, sizeof(*sin6));
sin6->sin6_len = sizeof(struct sockaddr_in6);
sin6->sin6_family = AF_INET6;
sin6->sin6_port = sin->sin_port;
sin6->sin6_addr.s6_addr32[0] = 0;
sin6->sin6_addr.s6_addr32[1] = 0;
sin6->sin6_addr.s6_addr32[2] = IPV6_ADDR_INT32_SMP;
sin6->sin6_addr.s6_addr32[3] = sin->sin_addr.s_addr;
}
/* Convert sockaddr_in6 into sockaddr_in. */
void
in6_sin6_2_sin_in_sock(struct sockaddr *nam)
{
struct sockaddr_in *sin_p;
struct sockaddr_in6 sin6;
/*
* Save original sockaddr_in6 addr and convert it
* to sockaddr_in.
*/
sin6 = *(struct sockaddr_in6 *)nam;
sin_p = (struct sockaddr_in *)nam;
in6_sin6_2_sin(sin_p, &sin6);
}
/* Convert sockaddr_in into sockaddr_in6 in v4 mapped addr format. */
void
in6_sin_2_v4mapsin6_in_sock(struct sockaddr **nam)
{
struct sockaddr_in *sin_p;
struct sockaddr_in6 *sin6_p;
sin6_p = malloc(sizeof *sin6_p, M_SONAME, M_WAITOK);
sin_p = (struct sockaddr_in *)*nam;
in6_sin_2_v4mapsin6(sin_p, sin6_p);
free(*nam, M_SONAME);
*nam = (struct sockaddr *)sin6_p;
}
diff --git a/sys/netinet6/in6.h b/sys/netinet6/in6.h
index 10bc1f91c554..34682da04898 100644
--- a/sys/netinet6/in6.h
+++ b/sys/netinet6/in6.h
@@ -1,755 +1,756 @@
/*-
* SPDX-License-Identifier: BSD-3-Clause
*
* 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: in6.h,v 1.89 2001/05/27 13:28:35 itojun Exp $
*/
/*-
* Copyright (c) 1982, 1986, 1990, 1993
* The Regents of the University of California. 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 University 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 REGENTS 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 REGENTS 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.
*
* @(#)in.h 8.3 (Berkeley) 1/3/94
* $FreeBSD$
*/
#ifndef __KAME_NETINET_IN_H_INCLUDED_
#error "do not include netinet6/in6.h directly, include netinet/in.h. see RFC2553"
#endif
#ifndef _NETINET6_IN6_H_
#define _NETINET6_IN6_H_
/*
* Identification of the network protocol stack
* for *BSD-current/release: http://www.kame.net/dev/cvsweb.cgi/kame/COVERAGE
* has the table of implementation/integration differences.
*/
#define __KAME__
#define __KAME_VERSION "FreeBSD"
/*
* IPv6 port allocation rules should mirror the IPv4 rules and are controlled
* by the net.inet.ip.portrange sysctl tree. The following defines exist
* for compatibility with userland applications that need them.
*/
#if __BSD_VISIBLE
#define IPV6PORT_RESERVED 1024
#define IPV6PORT_ANONMIN 49152
#define IPV6PORT_ANONMAX 65535
#define IPV6PORT_RESERVEDMIN 600
#define IPV6PORT_RESERVEDMAX (IPV6PORT_RESERVED-1)
#endif
/*
* IPv6 address
*/
struct in6_addr {
union {
uint8_t __u6_addr8[16];
uint16_t __u6_addr16[8];
uint32_t __u6_addr32[4];
} __u6_addr; /* 128-bit IP6 address */
};
#define s6_addr __u6_addr.__u6_addr8
#if defined(_KERNEL) || defined(_STANDALONE) /* XXX nonstandard */
#define s6_addr8 __u6_addr.__u6_addr8
#define s6_addr16 __u6_addr.__u6_addr16
#define s6_addr32 __u6_addr.__u6_addr32
#endif
#define INET6_ADDRSTRLEN 46
/*
* XXX missing POSIX.1-2001 macro IPPROTO_IPV6.
*/
/*
* Socket address for IPv6
*/
#if __BSD_VISIBLE
#define SIN6_LEN
#endif
struct sockaddr_in6 {
uint8_t sin6_len; /* length of this struct */
sa_family_t sin6_family; /* AF_INET6 */
in_port_t sin6_port; /* Transport layer port # */
uint32_t sin6_flowinfo; /* IP6 flow information */
struct in6_addr sin6_addr; /* IP6 address */
uint32_t sin6_scope_id; /* scope zone index */
};
/*
* Local definition for masks
*/
#ifdef _KERNEL /* XXX nonstandard */
#define IN6MASK0 {{{ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 }}}
#define IN6MASK32 {{{ 0xff, 0xff, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }}}
#define IN6MASK64 {{{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }}}
#define IN6MASK96 {{{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, \
0xff, 0xff, 0xff, 0xff, 0x00, 0x00, 0x00, 0x00 }}}
#define IN6MASK128 {{{ 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, \
0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff }}}
#endif
#ifdef _KERNEL
extern const struct sockaddr_in6 sa6_any;
extern const struct in6_addr in6mask0;
extern const struct in6_addr in6mask32;
extern const struct in6_addr in6mask64;
extern const struct in6_addr in6mask96;
extern const struct in6_addr in6mask128;
#endif /* _KERNEL */
/*
* Macros started with IPV6_ADDR is KAME local
*/
#ifdef _KERNEL /* XXX nonstandard */
#if _BYTE_ORDER == _BIG_ENDIAN
#define IPV6_ADDR_INT32_ONE 1
#define IPV6_ADDR_INT32_TWO 2
#define IPV6_ADDR_INT32_MNL 0xff010000
#define IPV6_ADDR_INT32_MLL 0xff020000
#define IPV6_ADDR_INT32_SMP 0x0000ffff
#define IPV6_ADDR_INT16_ULL 0xfe80
#define IPV6_ADDR_INT16_USL 0xfec0
#define IPV6_ADDR_INT16_MLL 0xff02
#elif _BYTE_ORDER == _LITTLE_ENDIAN
#define IPV6_ADDR_INT32_ONE 0x01000000
#define IPV6_ADDR_INT32_TWO 0x02000000
#define IPV6_ADDR_INT32_MNL 0x000001ff
#define IPV6_ADDR_INT32_MLL 0x000002ff
#define IPV6_ADDR_INT32_SMP 0xffff0000
#define IPV6_ADDR_INT16_ULL 0x80fe
#define IPV6_ADDR_INT16_USL 0xc0fe
#define IPV6_ADDR_INT16_MLL 0x02ff
#endif
#endif
/*
* Definition of some useful macros to handle IP6 addresses
*/
#if __BSD_VISIBLE
#define IN6ADDR_ANY_INIT \
{{{ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }}}
#define IN6ADDR_LOOPBACK_INIT \
{{{ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01 }}}
#define IN6ADDR_NODELOCAL_ALLNODES_INIT \
{{{ 0xff, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01 }}}
#define IN6ADDR_INTFACELOCAL_ALLNODES_INIT \
{{{ 0xff, 0x01, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01 }}}
#define IN6ADDR_LINKLOCAL_ALLNODES_INIT \
{{{ 0xff, 0x02, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01 }}}
#define IN6ADDR_LINKLOCAL_ALLROUTERS_INIT \
{{{ 0xff, 0x02, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x02 }}}
#define IN6ADDR_LINKLOCAL_ALLV2ROUTERS_INIT \
{{{ 0xff, 0x02, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, \
0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x16 }}}
#endif
extern const struct in6_addr in6addr_any;
extern const struct in6_addr in6addr_loopback;
#if __BSD_VISIBLE
extern const struct in6_addr in6addr_nodelocal_allnodes;
extern const struct in6_addr in6addr_linklocal_allnodes;
extern const struct in6_addr in6addr_linklocal_allrouters;
extern const struct in6_addr in6addr_linklocal_allv2routers;
#endif
/*
* Equality
* NOTE: Some of kernel programming environment (for example, openbsd/sparc)
* does not supply memcmp(). For userland memcmp() is preferred as it is
* in ANSI standard.
*/
#ifdef _KERNEL
#define IN6_ARE_ADDR_EQUAL(a, b) \
(bcmp(&(a)->s6_addr[0], &(b)->s6_addr[0], sizeof(struct in6_addr)) == 0)
#else
#if __BSD_VISIBLE
#define IN6_ARE_ADDR_EQUAL(a, b) \
(memcmp(&(a)->s6_addr[0], &(b)->s6_addr[0], sizeof(struct in6_addr)) == 0)
#endif
#endif
/*
* Unspecified
*/
#define IN6_IS_ADDR_UNSPECIFIED(a) \
((a)->__u6_addr.__u6_addr32[0] == 0 && \
(a)->__u6_addr.__u6_addr32[1] == 0 && \
(a)->__u6_addr.__u6_addr32[2] == 0 && \
(a)->__u6_addr.__u6_addr32[3] == 0)
/*
* Loopback
*/
#define IN6_IS_ADDR_LOOPBACK(a) \
((a)->__u6_addr.__u6_addr32[0] == 0 && \
(a)->__u6_addr.__u6_addr32[1] == 0 && \
(a)->__u6_addr.__u6_addr32[2] == 0 && \
(a)->__u6_addr.__u6_addr32[3] == ntohl(1))
/*
* IPv4 compatible
*/
#define IN6_IS_ADDR_V4COMPAT(a) \
((a)->__u6_addr.__u6_addr32[0] == 0 && \
(a)->__u6_addr.__u6_addr32[1] == 0 && \
(a)->__u6_addr.__u6_addr32[2] == 0 && \
(a)->__u6_addr.__u6_addr32[3] != 0 && \
(a)->__u6_addr.__u6_addr32[3] != ntohl(1))
/*
* Mapped
*/
#define IN6_IS_ADDR_V4MAPPED(a) \
((a)->__u6_addr.__u6_addr32[0] == 0 && \
(a)->__u6_addr.__u6_addr32[1] == 0 && \
(a)->__u6_addr.__u6_addr32[2] == ntohl(0x0000ffff))
/*
* KAME Scope Values
*/
#ifdef _KERNEL /* XXX nonstandard */
#define IPV6_ADDR_SCOPE_NODELOCAL 0x01
#define IPV6_ADDR_SCOPE_INTFACELOCAL 0x01
#define IPV6_ADDR_SCOPE_LINKLOCAL 0x02
#define IPV6_ADDR_SCOPE_SITELOCAL 0x05
#define IPV6_ADDR_SCOPE_ORGLOCAL 0x08 /* just used in this file */
#define IPV6_ADDR_SCOPE_GLOBAL 0x0e
#else
#define __IPV6_ADDR_SCOPE_NODELOCAL 0x01
#define __IPV6_ADDR_SCOPE_INTFACELOCAL 0x01
#define __IPV6_ADDR_SCOPE_LINKLOCAL 0x02
#define __IPV6_ADDR_SCOPE_SITELOCAL 0x05
#define __IPV6_ADDR_SCOPE_ORGLOCAL 0x08 /* just used in this file */
#define __IPV6_ADDR_SCOPE_GLOBAL 0x0e
#endif
/*
* Unicast Scope
* Note that we must check topmost 10 bits only, not 16 bits (see RFC2373).
*/
#define IN6_IS_ADDR_LINKLOCAL(a) \
(((a)->s6_addr[0] == 0xfe) && (((a)->s6_addr[1] & 0xc0) == 0x80))
#define IN6_IS_ADDR_SITELOCAL(a) \
(((a)->s6_addr[0] == 0xfe) && (((a)->s6_addr[1] & 0xc0) == 0xc0))
/*
* Multicast
*/
#define IN6_IS_ADDR_MULTICAST(a) ((a)->s6_addr[0] == 0xff)
#ifdef _KERNEL /* XXX nonstandard */
#define IPV6_ADDR_MC_SCOPE(a) ((a)->s6_addr[1] & 0x0f)
#else
#define __IPV6_ADDR_MC_SCOPE(a) ((a)->s6_addr[1] & 0x0f)
#endif
/*
* Multicast Scope
*/
#ifdef _KERNEL /* refers nonstandard items */
#define IN6_IS_ADDR_MC_NODELOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(IPV6_ADDR_MC_SCOPE(a) == IPV6_ADDR_SCOPE_NODELOCAL))
#define IN6_IS_ADDR_MC_INTFACELOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(IPV6_ADDR_MC_SCOPE(a) == IPV6_ADDR_SCOPE_INTFACELOCAL))
#define IN6_IS_ADDR_MC_LINKLOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(IPV6_ADDR_MC_SCOPE(a) == IPV6_ADDR_SCOPE_LINKLOCAL))
#define IN6_IS_ADDR_MC_SITELOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(IPV6_ADDR_MC_SCOPE(a) == IPV6_ADDR_SCOPE_SITELOCAL))
#define IN6_IS_ADDR_MC_ORGLOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(IPV6_ADDR_MC_SCOPE(a) == IPV6_ADDR_SCOPE_ORGLOCAL))
#define IN6_IS_ADDR_MC_GLOBAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(IPV6_ADDR_MC_SCOPE(a) == IPV6_ADDR_SCOPE_GLOBAL))
#else
#define IN6_IS_ADDR_MC_NODELOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(__IPV6_ADDR_MC_SCOPE(a) == __IPV6_ADDR_SCOPE_NODELOCAL))
#define IN6_IS_ADDR_MC_LINKLOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(__IPV6_ADDR_MC_SCOPE(a) == __IPV6_ADDR_SCOPE_LINKLOCAL))
#define IN6_IS_ADDR_MC_SITELOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(__IPV6_ADDR_MC_SCOPE(a) == __IPV6_ADDR_SCOPE_SITELOCAL))
#define IN6_IS_ADDR_MC_ORGLOCAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(__IPV6_ADDR_MC_SCOPE(a) == __IPV6_ADDR_SCOPE_ORGLOCAL))
#define IN6_IS_ADDR_MC_GLOBAL(a) \
(IN6_IS_ADDR_MULTICAST(a) && \
(__IPV6_ADDR_MC_SCOPE(a) == __IPV6_ADDR_SCOPE_GLOBAL))
#endif
#ifdef _KERNEL /* nonstandard */
/*
* KAME Scope
*/
#define IN6_IS_SCOPE_LINKLOCAL(a) \
((IN6_IS_ADDR_LINKLOCAL(a)) || \
(IN6_IS_ADDR_MC_LINKLOCAL(a)))
#define IN6_IS_SCOPE_EMBED(a) \
((IN6_IS_ADDR_LINKLOCAL(a)) || \
(IN6_IS_ADDR_MC_LINKLOCAL(a)) || \
(IN6_IS_ADDR_MC_INTFACELOCAL(a)))
#define IFA6_IS_DEPRECATED(a) \
((a)->ia6_lifetime.ia6t_pltime != ND6_INFINITE_LIFETIME && \
(u_int32_t)((time_uptime - (a)->ia6_updatetime)) > \
(a)->ia6_lifetime.ia6t_pltime)
#define IFA6_IS_INVALID(a) \
((a)->ia6_lifetime.ia6t_vltime != ND6_INFINITE_LIFETIME && \
(u_int32_t)((time_uptime - (a)->ia6_updatetime)) > \
(a)->ia6_lifetime.ia6t_vltime)
#endif /* _KERNEL */
/*
* IP6 route structure
*/
#if __BSD_VISIBLE
struct nhop_object;
struct route_in6 {
struct nhop_object *ro_nh;
struct llentry *ro_lle;
/*
* ro_prepend and ro_plen are only used for bpf to pass in a
* preformed header. They are not cacheable.
*/
char *ro_prepend;
uint16_t ro_plen;
uint16_t ro_flags;
uint16_t ro_mtu; /* saved ro_rt mtu */
uint16_t spare;
struct sockaddr_in6 ro_dst;
};
#endif
#ifdef _KERNEL
#define MTAG_ABI_IPV6 1444287380 /* IPv6 ABI */
#define IPV6_TAG_DIRECT 0 /* direct-dispatch IPv6 */
#endif /* _KERNEL */
/*
* Options for use with [gs]etsockopt at the IPV6 level.
* First word of comment is data type; bool is stored in int.
*/
/* no hdrincl */
#if 0 /* the followings are relic in IPv4 and hence are disabled */
#define IPV6_OPTIONS 1 /* buf/ip6_opts; set/get IP6 options */
#define IPV6_RECVOPTS 5 /* bool; receive all IP6 opts w/dgram */
#define IPV6_RECVRETOPTS 6 /* bool; receive IP6 opts for response */
#define IPV6_RECVDSTADDR 7 /* bool; receive IP6 dst addr w/dgram */
#define IPV6_RETOPTS 8 /* ip6_opts; set/get IP6 options */
#endif
#define IPV6_SOCKOPT_RESERVED1 3 /* reserved for future use */
#define IPV6_UNICAST_HOPS 4 /* int; IP6 hops */
#define IPV6_MULTICAST_IF 9 /* u_int; set/get IP6 multicast i/f */
#define IPV6_MULTICAST_HOPS 10 /* int; set/get IP6 multicast hops */
#define IPV6_MULTICAST_LOOP 11 /* u_int; set/get IP6 multicast loopback */
#define IPV6_JOIN_GROUP 12 /* ipv6_mreq; join a group membership */
#define IPV6_LEAVE_GROUP 13 /* ipv6_mreq; leave a group membership */
#define IPV6_PORTRANGE 14 /* int; range to choose for unspec port */
#define ICMP6_FILTER 18 /* icmp6_filter; icmp6 filter */
/* RFC2292 options */
#ifdef _KERNEL
#define IPV6_2292PKTINFO 19 /* bool; send/recv if, src/dst addr */
#define IPV6_2292HOPLIMIT 20 /* bool; hop limit */
#define IPV6_2292NEXTHOP 21 /* bool; next hop addr */
#define IPV6_2292HOPOPTS 22 /* bool; hop-by-hop option */
#define IPV6_2292DSTOPTS 23 /* bool; destinaion option */
#define IPV6_2292RTHDR 24 /* bool; routing header */
#define IPV6_2292PKTOPTIONS 25 /* buf/cmsghdr; set/get IPv6 options */
#endif
#define IPV6_CHECKSUM 26 /* int; checksum offset for raw socket */
#define IPV6_V6ONLY 27 /* bool; make AF_INET6 sockets v6 only */
#ifndef _KERNEL
#define IPV6_BINDV6ONLY IPV6_V6ONLY
#endif
#define IPV6_IPSEC_POLICY 28 /* struct; get/set security policy */
/* 29; unused; was IPV6_FAITH */
#if 1 /* IPV6FIREWALL */
#define IPV6_FW_ADD 30 /* add a firewall rule to chain */
#define IPV6_FW_DEL 31 /* delete a firewall rule from chain */
#define IPV6_FW_FLUSH 32 /* flush firewall rule chain */
#define IPV6_FW_ZERO 33 /* clear single/all firewall counter(s) */
#define IPV6_FW_GET 34 /* get entire firewall rule chain */
#endif
/* new socket options introduced in RFC3542 */
#define IPV6_RTHDRDSTOPTS 35 /* ip6_dest; send dst option before rthdr */
#define IPV6_RECVPKTINFO 36 /* bool; recv if, dst addr */
#define IPV6_RECVHOPLIMIT 37 /* bool; recv hop limit */
#define IPV6_RECVRTHDR 38 /* bool; recv routing header */
#define IPV6_RECVHOPOPTS 39 /* bool; recv hop-by-hop option */
#define IPV6_RECVDSTOPTS 40 /* bool; recv dst option after rthdr */
#ifdef _KERNEL
#define IPV6_RECVRTHDRDSTOPTS 41 /* bool; recv dst option before rthdr */
#endif
#define IPV6_USE_MIN_MTU 42 /* bool; send packets at the minimum MTU */
#define IPV6_RECVPATHMTU 43 /* bool; notify an according MTU */
#define IPV6_PATHMTU 44 /* mtuinfo; get the current path MTU (sopt),
4 bytes int; MTU notification (cmsg) */
#if 0 /*obsoleted during 2292bis -> 3542*/
#define IPV6_REACHCONF 45 /* no data; ND reachability confirm
(cmsg only/not in of RFC3542) */
#endif
/* more new socket options introduced in RFC3542 */
#define IPV6_PKTINFO 46 /* in6_pktinfo; send if, src addr */
#define IPV6_HOPLIMIT 47 /* int; send hop limit */
#define IPV6_NEXTHOP 48 /* sockaddr; next hop addr */
#define IPV6_HOPOPTS 49 /* ip6_hbh; send hop-by-hop option */
#define IPV6_DSTOPTS 50 /* ip6_dest; send dst option befor rthdr */
#define IPV6_RTHDR 51 /* ip6_rthdr; send routing header */
#if 0
#define IPV6_PKTOPTIONS 52 /* buf/cmsghdr; set/get IPv6 options */
/* obsoleted by RFC3542 */
#endif
#define IPV6_RECVTCLASS 57 /* bool; recv traffic class values */
#define IPV6_AUTOFLOWLABEL 59 /* bool; attach flowlabel automagically */
#define IPV6_TCLASS 61 /* int; send traffic class value */
#define IPV6_DONTFRAG 62 /* bool; disable IPv6 fragmentation */
#define IPV6_PREFER_TEMPADDR 63 /* int; prefer temporary addresses as
* the source address.
*/
#define IPV6_BINDANY 64 /* bool: allow bind to any address */
#define IPV6_BINDMULTI 65 /* bool; allow multibind to same addr/port */
#define IPV6_RSS_LISTEN_BUCKET 66 /* int; set RSS listen bucket */
#define IPV6_FLOWID 67 /* int; flowid of given socket */
#define IPV6_FLOWTYPE 68 /* int; flowtype of given socket */
#define IPV6_RSSBUCKETID 69 /* int; RSS bucket ID of given socket */
#define IPV6_RECVFLOWID 70 /* bool; receive IP6 flowid/flowtype w/ datagram */
#define IPV6_RECVRSSBUCKETID 71 /* bool; receive IP6 RSS bucket id w/ datagram */
#define IPV6_ORIGDSTADDR 72 /* bool: allow getting dstaddr /port info */
#define IPV6_RECVORIGDSTADDR IPV6_ORIGDSTADDR
/*
* The following option is private; do not use it from user applications.
* It is deliberately defined to the same value as IP_MSFILTER.
*/
#define IPV6_MSFILTER 74 /* struct __msfilterreq;
* set/get multicast source filter list.
*/
/* The following option deals with the 802.1Q Ethernet Priority Code Point */
#define IPV6_VLAN_PCP 75 /* int; set/get PCP used for packet, */
/* -1 use interface default */
/* to define items, should talk with KAME guys first, for *BSD compatibility */
#define IPV6_RTHDR_LOOSE 0 /* this hop need not be a neighbor. XXX old spec */
#define IPV6_RTHDR_STRICT 1 /* this hop must be a neighbor. XXX old spec */
#define IPV6_RTHDR_TYPE_0 0 /* IPv6 routing header type 0 */
/*
* Defaults and limits for options
*/
#define IPV6_DEFAULT_MULTICAST_HOPS 1 /* normally limit m'casts to 1 hop */
#define IPV6_DEFAULT_MULTICAST_LOOP 1 /* normally hear sends if a member */
/*
* Limit for IPv6 multicast memberships
*/
#define IPV6_MAX_MEMBERSHIPS 4095
/*
* Default resource limits for IPv6 multicast source filtering.
* These may be modified by sysctl.
*/
#define IPV6_MAX_GROUP_SRC_FILTER 512 /* sources per group */
#define IPV6_MAX_SOCK_SRC_FILTER 128 /* sources per socket/group */
/*
* Argument structure for IPV6_JOIN_GROUP and IPV6_LEAVE_GROUP.
*/
struct ipv6_mreq {
struct in6_addr ipv6mr_multiaddr;
unsigned int ipv6mr_interface;
};
/*
* IPV6_PKTINFO: Packet information(RFC2292 sec 5)
*/
struct in6_pktinfo {
struct in6_addr ipi6_addr; /* src/dst IPv6 address */
unsigned int ipi6_ifindex; /* send/recv interface index */
};
/*
* Control structure for IPV6_RECVPATHMTU socket option.
*/
struct ip6_mtuinfo {
struct sockaddr_in6 ip6m_addr; /* or sockaddr_storage? */
uint32_t ip6m_mtu;
};
/*
* Argument for IPV6_PORTRANGE:
* - which range to search when port is unspecified at bind() or connect()
*/
#define IPV6_PORTRANGE_DEFAULT 0 /* default range */
#define IPV6_PORTRANGE_HIGH 1 /* "high" - request firewall bypass */
#define IPV6_PORTRANGE_LOW 2 /* "low" - vouchsafe security */
#if __BSD_VISIBLE
/*
* Definitions for inet6 sysctl operations.
*
* Third level is protocol number.
* Fourth level is desired variable within that protocol.
*/
#define IPV6PROTO_MAXID (IPPROTO_PIM + 1) /* don't list to IPV6PROTO_MAX */
/*
* Names for IP sysctl objects
*/
#define IPV6CTL_FORWARDING 1 /* act as router */
#define IPV6CTL_SENDREDIRECTS 2 /* may send redirects when forwarding*/
#define IPV6CTL_DEFHLIM 3 /* default Hop-Limit */
#ifdef notyet
#define IPV6CTL_DEFMTU 4 /* default MTU */
#endif
#define IPV6CTL_FORWSRCRT 5 /* forward source-routed dgrams */
#define IPV6CTL_STATS 6 /* stats */
#define IPV6CTL_MRTSTATS 7 /* multicast forwarding stats */
#define IPV6CTL_MRTPROTO 8 /* multicast routing protocol */
#define IPV6CTL_MAXFRAGPACKETS 9 /* max packets reassembly queue */
#define IPV6CTL_SOURCECHECK 10 /* verify source route and intf */
#define IPV6CTL_SOURCECHECK_LOGINT 11 /* minimume logging interval */
#define IPV6CTL_ACCEPT_RTADV 12
/* 13; unused; was: IPV6CTL_KEEPFAITH */
#define IPV6CTL_LOG_INTERVAL 14
#define IPV6CTL_HDRNESTLIMIT 15
#define IPV6CTL_DAD_COUNT 16
#define IPV6CTL_AUTO_FLOWLABEL 17
#define IPV6CTL_DEFMCASTHLIM 18
#define IPV6CTL_GIF_HLIM 19 /* default HLIM for gif encap packet */
#define IPV6CTL_KAME_VERSION 20
#define IPV6CTL_USE_DEPRECATED 21 /* use deprecated addr (RFC2462 5.5.4) */
#define IPV6CTL_RR_PRUNE 22 /* walk timer for router renumbering */
#if 0 /* obsolete */
#define IPV6CTL_MAPPED_ADDR 23
#endif
#define IPV6CTL_V6ONLY 24
/* IPV6CTL_RTEXPIRE 25 deprecated */
/* IPV6CTL_RTMINEXPIRE 26 deprecated */
/* IPV6CTL_RTMAXCACHE 27 deprecated */
#define IPV6CTL_USETEMPADDR 32 /* use temporary addresses (RFC3041) */
#define IPV6CTL_TEMPPLTIME 33 /* preferred lifetime for tmpaddrs */
#define IPV6CTL_TEMPVLTIME 34 /* valid lifetime for tmpaddrs */
#define IPV6CTL_AUTO_LINKLOCAL 35 /* automatic link-local addr assign */
#define IPV6CTL_RIP6STATS 36 /* raw_ip6 stats */
#define IPV6CTL_PREFER_TEMPADDR 37 /* prefer temporary addr as src */
#define IPV6CTL_ADDRCTLPOLICY 38 /* get/set address selection policy */
#define IPV6CTL_USE_DEFAULTZONE 39 /* use default scope zone */
#define IPV6CTL_MAXFRAGS 41 /* max fragments */
#if 0
#define IPV6CTL_IFQ 42 /* ip6intrq node */
#define IPV6CTL_ISATAPRTR 43 /* isatap router */
#endif
#define IPV6CTL_MCAST_PMTU 44 /* enable pMTU discovery for multicast? */
/* New entries should be added here from current IPV6CTL_MAXID value. */
/* to define items, should talk with KAME guys first, for *BSD compatibility */
#define IPV6CTL_STEALTH 45
#define ICMPV6CTL_ND6_ONLINKNSRFC4861 47
#define IPV6CTL_NO_RADR 48 /* No defroute from RA */
#define IPV6CTL_NORBIT_RAIF 49 /* Disable R-bit in NA on RA
* receiving IF. */
#define IPV6CTL_RFC6204W3 50 /* Accept defroute even when forwarding
enabled */
#define IPV6CTL_INTRQMAXLEN 51 /* max length of IPv6 netisr queue */
#define IPV6CTL_INTRDQMAXLEN 52 /* max length of direct IPv6 netisr
* queue */
#define IPV6CTL_MAXFRAGSPERPACKET 53 /* Max fragments per packet */
#define IPV6CTL_MAXFRAGBUCKETSIZE 54 /* Max reassembly queues per bucket */
#define IPV6CTL_MAXID 55
#endif /* __BSD_VISIBLE */
/*
* Since both netinet/ and netinet6/ call into netipsec/ and netpfil/,
* the protocol specific mbuf flags are shared between them.
*/
#define M_FASTFWD_OURS M_PROTO1 /* changed dst to local */
#define M_IP6_NEXTHOP M_PROTO2 /* explicit ip nexthop */
#define M_IP_NEXTHOP M_PROTO2 /* explicit ip nexthop */
#define M_SKIP_FIREWALL M_PROTO3 /* skip firewall processing */
#define M_AUTHIPHDR M_PROTO4
#define M_DECRYPTED M_PROTO5
#define M_LOOP M_PROTO6
#define M_AUTHIPDGM M_PROTO7
#define M_RTALERT_MLD M_PROTO8
#define M_FRAGMENTED M_PROTO9 /* contained fragment header */
#ifdef _KERNEL
struct cmsghdr;
struct ip6_hdr;
int in6_cksum_pseudo(struct ip6_hdr *, uint32_t, uint8_t, uint16_t);
int in6_cksum(struct mbuf *, u_int8_t, u_int32_t, u_int32_t);
int in6_cksum_partial(struct mbuf *, u_int8_t, u_int32_t, u_int32_t,
u_int32_t);
int in6_localaddr(struct in6_addr *);
int in6_localip(struct in6_addr *);
+bool in6_localip_fib(struct in6_addr *, uint16_t);
int in6_ifhasaddr(struct ifnet *, struct in6_addr *);
int in6_addrscope(const struct in6_addr *);
char *ip6_sprintf(char *, const struct in6_addr *);
struct in6_ifaddr *in6_ifawithifp(struct ifnet *, struct in6_addr *);
extern void in6_if_up(struct ifnet *);
struct sockaddr;
extern u_char ip6_protox[];
void in6_sin6_2_sin(struct sockaddr_in *sin,
struct sockaddr_in6 *sin6);
void in6_sin_2_v4mapsin6(struct sockaddr_in *sin,
struct sockaddr_in6 *sin6);
void in6_sin6_2_sin_in_sock(struct sockaddr *nam);
void in6_sin_2_v4mapsin6_in_sock(struct sockaddr **nam);
extern void addrsel_policy_init(void);
#define satosin6(sa) ((struct sockaddr_in6 *)(sa))
#define sin6tosa(sin6) ((struct sockaddr *)(sin6))
#define ifatoia6(ifa) ((struct in6_ifaddr *)(ifa))
#endif /* _KERNEL */
#ifndef _SIZE_T_DECLARED
typedef __size_t size_t;
#define _SIZE_T_DECLARED
#endif
#ifndef _SOCKLEN_T_DECLARED
typedef __socklen_t socklen_t;
#define _SOCKLEN_T_DECLARED
#endif
#if __BSD_VISIBLE
__BEGIN_DECLS
struct cmsghdr;
extern int inet6_option_space(int);
extern int inet6_option_init(void *, struct cmsghdr **, int);
extern int inet6_option_append(struct cmsghdr *, const uint8_t *,
int, int);
extern uint8_t *inet6_option_alloc(struct cmsghdr *, int, int, int);
extern int inet6_option_next(const struct cmsghdr *, uint8_t **);
extern int inet6_option_find(const struct cmsghdr *, uint8_t **, int);
extern size_t inet6_rthdr_space(int, int);
extern struct cmsghdr *inet6_rthdr_init(void *, int);
extern int inet6_rthdr_add(struct cmsghdr *, const struct in6_addr *,
unsigned int);
extern int inet6_rthdr_lasthop(struct cmsghdr *, unsigned int);
#if 0 /* not implemented yet */
extern int inet6_rthdr_reverse(const struct cmsghdr *, struct cmsghdr *);
#endif
extern int inet6_rthdr_segments(const struct cmsghdr *);
extern struct in6_addr *inet6_rthdr_getaddr(struct cmsghdr *, int);
extern int inet6_rthdr_getflags(const struct cmsghdr *, int);
extern int inet6_opt_init(void *, socklen_t);
extern int inet6_opt_append(void *, socklen_t, int, uint8_t, socklen_t,
uint8_t, void **);
extern int inet6_opt_finish(void *, socklen_t, int);
extern int inet6_opt_set_val(void *, int, void *, socklen_t);
extern int inet6_opt_next(void *, socklen_t, int, uint8_t *, socklen_t *,
void **);
extern int inet6_opt_find(void *, socklen_t, int, uint8_t, socklen_t *,
void **);
extern int inet6_opt_get_val(void *, int, void *, socklen_t);
extern socklen_t inet6_rth_space(int, int);
extern void *inet6_rth_init(void *, socklen_t, int, int);
extern int inet6_rth_add(void *, const struct in6_addr *);
extern int inet6_rth_reverse(const void *, void *);
extern int inet6_rth_segments(const void *);
extern struct in6_addr *inet6_rth_getaddr(const void *, int);
__END_DECLS
#endif /* __BSD_VISIBLE */
#endif /* !_NETINET6_IN6_H_ */