If source of ARP request didn't pass the routing check (e.g. not in directly connected network), be polite and still answer the request instead of dropping frame. Reported by: quadro at irc@rusnet
1299 lines
35 KiB
C
1299 lines
35 KiB
C
/*-
|
|
* Copyright (c) 1982, 1986, 1988, 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.
|
|
* 4. 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.
|
|
*
|
|
* @(#)if_ether.c 8.1 (Berkeley) 6/10/93
|
|
*/
|
|
|
|
/*
|
|
* Ethernet address resolution protocol.
|
|
* TODO:
|
|
* add "inuse/lock" bit (or ref. count) along with valid bit
|
|
*/
|
|
|
|
#include <sys/cdefs.h>
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
#include "opt_inet.h"
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/kernel.h>
|
|
#include <sys/lock.h>
|
|
#include <sys/queue.h>
|
|
#include <sys/sysctl.h>
|
|
#include <sys/systm.h>
|
|
#include <sys/mbuf.h>
|
|
#include <sys/malloc.h>
|
|
#include <sys/proc.h>
|
|
#include <sys/rmlock.h>
|
|
#include <sys/socket.h>
|
|
#include <sys/syslog.h>
|
|
|
|
#include <net/if.h>
|
|
#include <net/if_var.h>
|
|
#include <net/if_dl.h>
|
|
#include <net/if_types.h>
|
|
#include <net/netisr.h>
|
|
#include <net/ethernet.h>
|
|
#include <net/route.h>
|
|
#include <net/vnet.h>
|
|
|
|
#include <netinet/in.h>
|
|
#include <netinet/in_fib.h>
|
|
#include <netinet/in_var.h>
|
|
#include <net/if_llatbl.h>
|
|
#include <netinet/if_ether.h>
|
|
#ifdef INET
|
|
#include <netinet/ip_carp.h>
|
|
#endif
|
|
|
|
#include <security/mac/mac_framework.h>
|
|
|
|
#define SIN(s) ((const struct sockaddr_in *)(s))
|
|
|
|
static struct timeval arp_lastlog;
|
|
static int arp_curpps;
|
|
static int arp_maxpps = 1;
|
|
|
|
/* Simple ARP state machine */
|
|
enum arp_llinfo_state {
|
|
ARP_LLINFO_INCOMPLETE = 0, /* No LLE data */
|
|
ARP_LLINFO_REACHABLE, /* LLE is valid */
|
|
ARP_LLINFO_VERIFY, /* LLE is valid, need refresh */
|
|
ARP_LLINFO_DELETED, /* LLE is deleted */
|
|
};
|
|
|
|
SYSCTL_DECL(_net_link_ether);
|
|
static SYSCTL_NODE(_net_link_ether, PF_INET, inet, CTLFLAG_RW, 0, "");
|
|
static SYSCTL_NODE(_net_link_ether, PF_ARP, arp, CTLFLAG_RW, 0, "");
|
|
|
|
/* timer values */
|
|
static VNET_DEFINE(int, arpt_keep) = (20*60); /* once resolved, good for 20
|
|
* minutes */
|
|
static VNET_DEFINE(int, arp_maxtries) = 5;
|
|
static VNET_DEFINE(int, arp_proxyall) = 0;
|
|
static VNET_DEFINE(int, arpt_down) = 20; /* keep incomplete entries for
|
|
* 20 seconds */
|
|
static VNET_DEFINE(int, arpt_rexmit) = 1; /* retransmit arp entries, sec*/
|
|
VNET_PCPUSTAT_DEFINE(struct arpstat, arpstat); /* ARP statistics, see if_arp.h */
|
|
VNET_PCPUSTAT_SYSINIT(arpstat);
|
|
|
|
#ifdef VIMAGE
|
|
VNET_PCPUSTAT_SYSUNINIT(arpstat);
|
|
#endif /* VIMAGE */
|
|
|
|
static VNET_DEFINE(int, arp_maxhold) = 1;
|
|
static VNET_DEFINE(int, arp_on_link) = 1;
|
|
|
|
#define V_arpt_keep VNET(arpt_keep)
|
|
#define V_arpt_down VNET(arpt_down)
|
|
#define V_arpt_rexmit VNET(arpt_rexmit)
|
|
#define V_arp_maxtries VNET(arp_maxtries)
|
|
#define V_arp_proxyall VNET(arp_proxyall)
|
|
#define V_arp_maxhold VNET(arp_maxhold)
|
|
#define V_arp_on_link VNET(arp_on_link)
|
|
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, max_age, CTLFLAG_VNET | CTLFLAG_RW,
|
|
&VNET_NAME(arpt_keep), 0,
|
|
"ARP entry lifetime in seconds");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, maxtries, CTLFLAG_VNET | CTLFLAG_RW,
|
|
&VNET_NAME(arp_maxtries), 0,
|
|
"ARP resolution attempts before returning error");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, proxyall, CTLFLAG_VNET | CTLFLAG_RW,
|
|
&VNET_NAME(arp_proxyall), 0,
|
|
"Enable proxy ARP for all suitable requests");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, wait, CTLFLAG_VNET | CTLFLAG_RW,
|
|
&VNET_NAME(arpt_down), 0,
|
|
"Incomplete ARP entry lifetime in seconds");
|
|
SYSCTL_VNET_PCPUSTAT(_net_link_ether_arp, OID_AUTO, stats, struct arpstat,
|
|
arpstat, "ARP statistics (struct arpstat, net/if_arp.h)");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, maxhold, CTLFLAG_VNET | CTLFLAG_RW,
|
|
&VNET_NAME(arp_maxhold), 0,
|
|
"Number of packets to hold per ARP entry");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, max_log_per_second,
|
|
CTLFLAG_RW, &arp_maxpps, 0,
|
|
"Maximum number of remotely triggered ARP messages that can be "
|
|
"logged per second");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, arp_on_link, CTLFLAG_VNET | CTLFLAG_RW,
|
|
&VNET_NAME(arp_on_link), 0,
|
|
"Send gratuitous ARP's on interface link up events");
|
|
|
|
#define ARP_LOG(pri, ...) do { \
|
|
if (ppsratecheck(&arp_lastlog, &arp_curpps, arp_maxpps)) \
|
|
log((pri), "arp: " __VA_ARGS__); \
|
|
} while (0)
|
|
|
|
|
|
static void arp_init(void);
|
|
static void arpintr(struct mbuf *);
|
|
static void arptimer(void *);
|
|
#ifdef INET
|
|
static void in_arpinput(struct mbuf *);
|
|
#endif
|
|
|
|
static void arp_check_update_lle(struct arphdr *ah, struct in_addr isaddr,
|
|
struct ifnet *ifp, int bridged, struct llentry *la);
|
|
static void arp_mark_lle_reachable(struct llentry *la);
|
|
static void arp_iflladdr(void *arg __unused, struct ifnet *ifp);
|
|
|
|
static eventhandler_tag iflladdr_tag;
|
|
static eventhandler_tag ifnet_link_event_tag;
|
|
|
|
static const struct netisr_handler arp_nh = {
|
|
.nh_name = "arp",
|
|
.nh_handler = arpintr,
|
|
.nh_proto = NETISR_ARP,
|
|
.nh_policy = NETISR_POLICY_SOURCE,
|
|
};
|
|
|
|
/*
|
|
* Timeout routine. Age arp_tab entries periodically.
|
|
*/
|
|
static void
|
|
arptimer(void *arg)
|
|
{
|
|
struct llentry *lle = (struct llentry *)arg;
|
|
struct ifnet *ifp;
|
|
int r_skip_req;
|
|
|
|
if (lle->la_flags & LLE_STATIC) {
|
|
return;
|
|
}
|
|
LLE_WLOCK(lle);
|
|
if (callout_pending(&lle->lle_timer)) {
|
|
/*
|
|
* Here we are a bit odd here in the treatment of
|
|
* active/pending. If the pending bit is set, it got
|
|
* rescheduled before I ran. The active
|
|
* bit we ignore, since if it was stopped
|
|
* in ll_tablefree() and was currently running
|
|
* it would have return 0 so the code would
|
|
* not have deleted it since the callout could
|
|
* not be stopped so we want to go through
|
|
* with the delete here now. If the callout
|
|
* was restarted, the pending bit will be back on and
|
|
* we just want to bail since the callout_reset would
|
|
* return 1 and our reference would have been removed
|
|
* by arpresolve() below.
|
|
*/
|
|
LLE_WUNLOCK(lle);
|
|
return;
|
|
}
|
|
ifp = lle->lle_tbl->llt_ifp;
|
|
CURVNET_SET(ifp->if_vnet);
|
|
|
|
switch (lle->ln_state) {
|
|
case ARP_LLINFO_REACHABLE:
|
|
|
|
/*
|
|
* Expiration time is approaching.
|
|
* Let's try to refresh entry if it is still
|
|
* in use.
|
|
*
|
|
* Set r_skip_req to get feedback from
|
|
* fast path. Change state and re-schedule
|
|
* ourselves.
|
|
*/
|
|
LLE_REQ_LOCK(lle);
|
|
lle->r_skip_req = 1;
|
|
LLE_REQ_UNLOCK(lle);
|
|
lle->ln_state = ARP_LLINFO_VERIFY;
|
|
callout_schedule(&lle->lle_timer, hz * V_arpt_rexmit);
|
|
LLE_WUNLOCK(lle);
|
|
CURVNET_RESTORE();
|
|
return;
|
|
case ARP_LLINFO_VERIFY:
|
|
LLE_REQ_LOCK(lle);
|
|
r_skip_req = lle->r_skip_req;
|
|
LLE_REQ_UNLOCK(lle);
|
|
|
|
if (r_skip_req == 0 && lle->la_preempt > 0) {
|
|
/* Entry was used, issue refresh request */
|
|
struct in_addr dst;
|
|
dst = lle->r_l3addr.addr4;
|
|
lle->la_preempt--;
|
|
callout_schedule(&lle->lle_timer, hz * V_arpt_rexmit);
|
|
LLE_WUNLOCK(lle);
|
|
arprequest(ifp, NULL, &dst, NULL);
|
|
CURVNET_RESTORE();
|
|
return;
|
|
}
|
|
/* Nothing happened. Reschedule if not too late */
|
|
if (lle->la_expire > time_uptime) {
|
|
callout_schedule(&lle->lle_timer, hz * V_arpt_rexmit);
|
|
LLE_WUNLOCK(lle);
|
|
CURVNET_RESTORE();
|
|
return;
|
|
}
|
|
break;
|
|
case ARP_LLINFO_INCOMPLETE:
|
|
case ARP_LLINFO_DELETED:
|
|
break;
|
|
}
|
|
|
|
if ((lle->la_flags & LLE_DELETED) == 0) {
|
|
int evt;
|
|
|
|
if (lle->la_flags & LLE_VALID)
|
|
evt = LLENTRY_EXPIRED;
|
|
else
|
|
evt = LLENTRY_TIMEDOUT;
|
|
EVENTHANDLER_INVOKE(lle_event, lle, evt);
|
|
}
|
|
|
|
callout_stop(&lle->lle_timer);
|
|
|
|
/* XXX: LOR avoidance. We still have ref on lle. */
|
|
LLE_WUNLOCK(lle);
|
|
IF_AFDATA_LOCK(ifp);
|
|
LLE_WLOCK(lle);
|
|
|
|
/* Guard against race with other llentry_free(). */
|
|
if (lle->la_flags & LLE_LINKED) {
|
|
LLE_REMREF(lle);
|
|
lltable_unlink_entry(lle->lle_tbl, lle);
|
|
}
|
|
IF_AFDATA_UNLOCK(ifp);
|
|
|
|
size_t pkts_dropped = llentry_free(lle);
|
|
|
|
ARPSTAT_ADD(dropped, pkts_dropped);
|
|
ARPSTAT_INC(timeouts);
|
|
|
|
CURVNET_RESTORE();
|
|
}
|
|
|
|
/*
|
|
* Broadcast an ARP request. Caller specifies:
|
|
* - arp header source ip address
|
|
* - arp header target ip address
|
|
* - arp header source ethernet address
|
|
*/
|
|
void
|
|
arprequest(struct ifnet *ifp, const struct in_addr *sip,
|
|
const struct in_addr *tip, u_char *enaddr)
|
|
{
|
|
struct mbuf *m;
|
|
struct arphdr *ah;
|
|
struct sockaddr sa;
|
|
u_char *carpaddr = NULL;
|
|
|
|
if (sip == NULL) {
|
|
/*
|
|
* The caller did not supply a source address, try to find
|
|
* a compatible one among those assigned to this interface.
|
|
*/
|
|
struct ifaddr *ifa;
|
|
|
|
IF_ADDR_RLOCK(ifp);
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
|
continue;
|
|
|
|
if (ifa->ifa_carp) {
|
|
if ((*carp_iamatch_p)(ifa, &carpaddr) == 0)
|
|
continue;
|
|
sip = &IA_SIN(ifa)->sin_addr;
|
|
} else {
|
|
carpaddr = NULL;
|
|
sip = &IA_SIN(ifa)->sin_addr;
|
|
}
|
|
|
|
if (0 == ((sip->s_addr ^ tip->s_addr) &
|
|
IA_MASKSIN(ifa)->sin_addr.s_addr))
|
|
break; /* found it. */
|
|
}
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
if (sip == NULL) {
|
|
printf("%s: cannot find matching address\n", __func__);
|
|
return;
|
|
}
|
|
}
|
|
if (enaddr == NULL)
|
|
enaddr = carpaddr ? carpaddr : (u_char *)IF_LLADDR(ifp);
|
|
|
|
if ((m = m_gethdr(M_NOWAIT, MT_DATA)) == NULL)
|
|
return;
|
|
m->m_len = sizeof(*ah) + 2 * sizeof(struct in_addr) +
|
|
2 * ifp->if_addrlen;
|
|
m->m_pkthdr.len = m->m_len;
|
|
M_ALIGN(m, m->m_len);
|
|
ah = mtod(m, struct arphdr *);
|
|
bzero((caddr_t)ah, m->m_len);
|
|
#ifdef MAC
|
|
mac_netinet_arp_send(ifp, m);
|
|
#endif
|
|
ah->ar_pro = htons(ETHERTYPE_IP);
|
|
ah->ar_hln = ifp->if_addrlen; /* hardware address length */
|
|
ah->ar_pln = sizeof(struct in_addr); /* protocol address length */
|
|
ah->ar_op = htons(ARPOP_REQUEST);
|
|
bcopy(enaddr, ar_sha(ah), ah->ar_hln);
|
|
bcopy(sip, ar_spa(ah), ah->ar_pln);
|
|
bcopy(tip, ar_tpa(ah), ah->ar_pln);
|
|
sa.sa_family = AF_ARP;
|
|
sa.sa_len = 2;
|
|
m->m_flags |= M_BCAST;
|
|
m_clrprotoflags(m); /* Avoid confusing lower layers. */
|
|
(*ifp->if_output)(ifp, m, &sa, NULL);
|
|
ARPSTAT_INC(txrequests);
|
|
}
|
|
|
|
/*
|
|
* Resolve an IP address into an ethernet address - heavy version.
|
|
* Used internally by arpresolve().
|
|
* We have already checked than we can't use existing lle without
|
|
* modification so we have to acquire LLE_EXCLUSIVE lle lock.
|
|
*
|
|
* On success, desten and flags are filled in and the function returns 0;
|
|
* If the packet must be held pending resolution, we return EWOULDBLOCK
|
|
* On other errors, we return the corresponding error code.
|
|
* Note that m_freem() handles NULL.
|
|
*/
|
|
static int
|
|
arpresolve_full(struct ifnet *ifp, int is_gw, int create, struct mbuf *m,
|
|
const struct sockaddr *dst, u_char *desten, uint32_t *pflags)
|
|
{
|
|
struct llentry *la = NULL, *la_tmp;
|
|
struct mbuf *curr = NULL;
|
|
struct mbuf *next = NULL;
|
|
int error, renew;
|
|
|
|
if (pflags != NULL)
|
|
*pflags = 0;
|
|
|
|
if (create == 0) {
|
|
IF_AFDATA_RLOCK(ifp);
|
|
la = lla_lookup(LLTABLE(ifp), LLE_EXCLUSIVE, dst);
|
|
IF_AFDATA_RUNLOCK(ifp);
|
|
}
|
|
if (la == NULL && (ifp->if_flags & (IFF_NOARP | IFF_STATICARP)) == 0) {
|
|
la = lltable_alloc_entry(LLTABLE(ifp), 0, dst);
|
|
if (la == NULL) {
|
|
log(LOG_DEBUG,
|
|
"arpresolve: can't allocate llinfo for %s on %s\n",
|
|
inet_ntoa(SIN(dst)->sin_addr), if_name(ifp));
|
|
m_freem(m);
|
|
return (EINVAL);
|
|
}
|
|
|
|
IF_AFDATA_WLOCK(ifp);
|
|
LLE_WLOCK(la);
|
|
la_tmp = lla_lookup(LLTABLE(ifp), LLE_EXCLUSIVE, dst);
|
|
/* Prefer ANY existing lle over newly-created one */
|
|
if (la_tmp == NULL)
|
|
lltable_link_entry(LLTABLE(ifp), la);
|
|
IF_AFDATA_WUNLOCK(ifp);
|
|
if (la_tmp != NULL) {
|
|
lltable_free_entry(LLTABLE(ifp), la);
|
|
la = la_tmp;
|
|
}
|
|
}
|
|
if (la == NULL) {
|
|
m_freem(m);
|
|
return (EINVAL);
|
|
}
|
|
|
|
if ((la->la_flags & LLE_VALID) &&
|
|
((la->la_flags & LLE_STATIC) || la->la_expire > time_uptime)) {
|
|
bcopy(&la->ll_addr, desten, ifp->if_addrlen);
|
|
|
|
/* Check if we have feedback request from arptimer() */
|
|
if (la->r_skip_req != 0) {
|
|
LLE_REQ_LOCK(la);
|
|
la->r_skip_req = 0; /* Notify that entry was used */
|
|
LLE_REQ_UNLOCK(la);
|
|
}
|
|
if (pflags != NULL)
|
|
*pflags = la->la_flags & (LLE_VALID|LLE_IFADDR);
|
|
LLE_WUNLOCK(la);
|
|
return (0);
|
|
}
|
|
|
|
renew = (la->la_asked == 0 || la->la_expire != time_uptime);
|
|
/*
|
|
* There is an arptab entry, but no ethernet address
|
|
* response yet. Add the mbuf to the list, dropping
|
|
* the oldest packet if we have exceeded the system
|
|
* setting.
|
|
*/
|
|
if (m != NULL) {
|
|
if (la->la_numheld >= V_arp_maxhold) {
|
|
if (la->la_hold != NULL) {
|
|
next = la->la_hold->m_nextpkt;
|
|
m_freem(la->la_hold);
|
|
la->la_hold = next;
|
|
la->la_numheld--;
|
|
ARPSTAT_INC(dropped);
|
|
}
|
|
}
|
|
if (la->la_hold != NULL) {
|
|
curr = la->la_hold;
|
|
while (curr->m_nextpkt != NULL)
|
|
curr = curr->m_nextpkt;
|
|
curr->m_nextpkt = m;
|
|
} else
|
|
la->la_hold = m;
|
|
la->la_numheld++;
|
|
}
|
|
/*
|
|
* Return EWOULDBLOCK if we have tried less than arp_maxtries. It
|
|
* will be masked by ether_output(). Return EHOSTDOWN/EHOSTUNREACH
|
|
* if we have already sent arp_maxtries ARP requests. Retransmit the
|
|
* ARP request, but not faster than one request per second.
|
|
*/
|
|
if (la->la_asked < V_arp_maxtries)
|
|
error = EWOULDBLOCK; /* First request. */
|
|
else
|
|
error = is_gw != 0 ? EHOSTUNREACH : EHOSTDOWN;
|
|
|
|
if (renew) {
|
|
int canceled;
|
|
|
|
LLE_ADDREF(la);
|
|
la->la_expire = time_uptime;
|
|
canceled = callout_reset(&la->lle_timer, hz * V_arpt_down,
|
|
arptimer, la);
|
|
if (canceled)
|
|
LLE_REMREF(la);
|
|
la->la_asked++;
|
|
LLE_WUNLOCK(la);
|
|
arprequest(ifp, NULL, &SIN(dst)->sin_addr, NULL);
|
|
return (error);
|
|
}
|
|
|
|
LLE_WUNLOCK(la);
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* Resolve an IP address into an ethernet address.
|
|
* On input:
|
|
* ifp is the interface we use
|
|
* is_gw != 0 if @dst represents gateway to some destination
|
|
* m is the mbuf. May be NULL if we don't have a packet.
|
|
* dst is the next hop,
|
|
* desten is the storage to put LL address.
|
|
* flags returns subset of lle flags: LLE_VALID | LLE_IFADDR
|
|
*
|
|
* On success, desten and flags are filled in and the function returns 0;
|
|
* If the packet must be held pending resolution, we return EWOULDBLOCK
|
|
* On other errors, we return the corresponding error code.
|
|
* Note that m_freem() handles NULL.
|
|
*/
|
|
int
|
|
arpresolve(struct ifnet *ifp, int is_gw, struct mbuf *m,
|
|
const struct sockaddr *dst, u_char *desten, uint32_t *pflags)
|
|
{
|
|
struct llentry *la = 0;
|
|
|
|
if (pflags != NULL)
|
|
*pflags = 0;
|
|
|
|
if (m != NULL) {
|
|
if (m->m_flags & M_BCAST) {
|
|
/* broadcast */
|
|
(void)memcpy(desten,
|
|
ifp->if_broadcastaddr, ifp->if_addrlen);
|
|
return (0);
|
|
}
|
|
if (m->m_flags & M_MCAST) {
|
|
/* multicast */
|
|
ETHER_MAP_IP_MULTICAST(&SIN(dst)->sin_addr, desten);
|
|
return (0);
|
|
}
|
|
}
|
|
|
|
IF_AFDATA_RLOCK(ifp);
|
|
la = lla_lookup(LLTABLE(ifp), LLE_UNLOCKED, dst);
|
|
if (la != NULL && (la->r_flags & RLLE_VALID) != 0) {
|
|
/* Entry found, let's copy lle info */
|
|
bcopy(&la->ll_addr, desten, ifp->if_addrlen);
|
|
if (pflags != NULL)
|
|
*pflags = LLE_VALID | (la->r_flags & RLLE_IFADDR);
|
|
/* Check if we have feedback request from arptimer() */
|
|
if (la->r_skip_req != 0) {
|
|
LLE_REQ_LOCK(la);
|
|
la->r_skip_req = 0; /* Notify that entry was used */
|
|
LLE_REQ_UNLOCK(la);
|
|
}
|
|
IF_AFDATA_RUNLOCK(ifp);
|
|
return (0);
|
|
}
|
|
IF_AFDATA_RUNLOCK(ifp);
|
|
|
|
return (arpresolve_full(ifp, is_gw, 1, m, dst, desten, pflags));
|
|
}
|
|
|
|
/*
|
|
* Common length and type checks are done here,
|
|
* then the protocol-specific routine is called.
|
|
*/
|
|
static void
|
|
arpintr(struct mbuf *m)
|
|
{
|
|
struct arphdr *ar;
|
|
struct ifnet *ifp;
|
|
char *layer;
|
|
int hlen;
|
|
|
|
ifp = m->m_pkthdr.rcvif;
|
|
|
|
if (m->m_len < sizeof(struct arphdr) &&
|
|
((m = m_pullup(m, sizeof(struct arphdr))) == NULL)) {
|
|
ARP_LOG(LOG_NOTICE, "packet with short header received on %s\n",
|
|
if_name(ifp));
|
|
return;
|
|
}
|
|
ar = mtod(m, struct arphdr *);
|
|
|
|
/* Check if length is sufficient */
|
|
if (m->m_len < arphdr_len(ar)) {
|
|
m = m_pullup(m, arphdr_len(ar));
|
|
if (m == NULL) {
|
|
ARP_LOG(LOG_NOTICE, "short packet received on %s\n",
|
|
if_name(ifp));
|
|
return;
|
|
}
|
|
ar = mtod(m, struct arphdr *);
|
|
}
|
|
|
|
hlen = 0;
|
|
layer = "";
|
|
switch (ntohs(ar->ar_hrd)) {
|
|
case ARPHRD_ETHER:
|
|
hlen = ETHER_ADDR_LEN; /* RFC 826 */
|
|
layer = "ethernet";
|
|
break;
|
|
case ARPHRD_IEEE802:
|
|
hlen = 6; /* RFC 1390, FDDI_ADDR_LEN */
|
|
layer = "fddi";
|
|
break;
|
|
case ARPHRD_ARCNET:
|
|
hlen = 1; /* RFC 1201, ARC_ADDR_LEN */
|
|
layer = "arcnet";
|
|
break;
|
|
case ARPHRD_INFINIBAND:
|
|
hlen = 20; /* RFC 4391, INFINIBAND_ALEN */
|
|
layer = "infiniband";
|
|
break;
|
|
case ARPHRD_IEEE1394:
|
|
hlen = 0; /* SHALL be 16 */ /* RFC 2734 */
|
|
layer = "firewire";
|
|
|
|
/*
|
|
* Restrict too long harware addresses.
|
|
* Currently we are capable of handling 20-byte
|
|
* addresses ( sizeof(lle->ll_addr) )
|
|
*/
|
|
if (ar->ar_hln >= 20)
|
|
hlen = 16;
|
|
break;
|
|
default:
|
|
ARP_LOG(LOG_NOTICE,
|
|
"packet with unknown harware format 0x%02d received on %s\n",
|
|
ntohs(ar->ar_hrd), if_name(ifp));
|
|
m_freem(m);
|
|
return;
|
|
}
|
|
|
|
if (hlen != 0 && hlen != ar->ar_hln) {
|
|
ARP_LOG(LOG_NOTICE,
|
|
"packet with invalid %s address length %d received on %s\n",
|
|
layer, ar->ar_hln, if_name(ifp));
|
|
m_freem(m);
|
|
return;
|
|
}
|
|
|
|
ARPSTAT_INC(received);
|
|
switch (ntohs(ar->ar_pro)) {
|
|
#ifdef INET
|
|
case ETHERTYPE_IP:
|
|
in_arpinput(m);
|
|
return;
|
|
#endif
|
|
}
|
|
m_freem(m);
|
|
}
|
|
|
|
#ifdef INET
|
|
/*
|
|
* ARP for Internet protocols on 10 Mb/s Ethernet.
|
|
* Algorithm is that given in RFC 826.
|
|
* In addition, a sanity check is performed on the sender
|
|
* protocol address, to catch impersonators.
|
|
* We no longer handle negotiations for use of trailer protocol:
|
|
* Formerly, ARP replied for protocol type ETHERTYPE_TRAIL sent
|
|
* along with IP replies if we wanted trailers sent to us,
|
|
* and also sent them in response to IP replies.
|
|
* This allowed either end to announce the desire to receive
|
|
* trailer packets.
|
|
* We no longer reply to requests for ETHERTYPE_TRAIL protocol either,
|
|
* but formerly didn't normally send requests.
|
|
*/
|
|
static int log_arp_wrong_iface = 1;
|
|
static int log_arp_movements = 1;
|
|
static int log_arp_permanent_modify = 1;
|
|
static int allow_multicast = 0;
|
|
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, log_arp_wrong_iface, CTLFLAG_RW,
|
|
&log_arp_wrong_iface, 0,
|
|
"log arp packets arriving on the wrong interface");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, log_arp_movements, CTLFLAG_RW,
|
|
&log_arp_movements, 0,
|
|
"log arp replies from MACs different than the one in the cache");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, log_arp_permanent_modify, CTLFLAG_RW,
|
|
&log_arp_permanent_modify, 0,
|
|
"log arp replies from MACs different than the one in the permanent arp entry");
|
|
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, allow_multicast, CTLFLAG_RW,
|
|
&allow_multicast, 0, "accept multicast addresses");
|
|
|
|
static void
|
|
in_arpinput(struct mbuf *m)
|
|
{
|
|
struct rm_priotracker in_ifa_tracker;
|
|
struct arphdr *ah;
|
|
struct ifnet *ifp = m->m_pkthdr.rcvif;
|
|
struct llentry *la = NULL, *la_tmp;
|
|
struct ifaddr *ifa;
|
|
struct in_ifaddr *ia;
|
|
struct sockaddr sa;
|
|
struct in_addr isaddr, itaddr, myaddr;
|
|
u_int8_t *enaddr = NULL;
|
|
int op;
|
|
int bridged = 0, is_bridge = 0;
|
|
int carped;
|
|
struct sockaddr_in sin;
|
|
struct sockaddr *dst;
|
|
struct nhop4_basic nh4;
|
|
|
|
sin.sin_len = sizeof(struct sockaddr_in);
|
|
sin.sin_family = AF_INET;
|
|
sin.sin_addr.s_addr = 0;
|
|
|
|
if (ifp->if_bridge)
|
|
bridged = 1;
|
|
if (ifp->if_type == IFT_BRIDGE)
|
|
is_bridge = 1;
|
|
|
|
/*
|
|
* We already have checked that mbuf contains enough contiguous data
|
|
* to hold entire arp message according to the arp header.
|
|
*/
|
|
ah = mtod(m, struct arphdr *);
|
|
|
|
/*
|
|
* ARP is only for IPv4 so we can reject packets with
|
|
* a protocol length not equal to an IPv4 address.
|
|
*/
|
|
if (ah->ar_pln != sizeof(struct in_addr)) {
|
|
ARP_LOG(LOG_NOTICE, "requested protocol length != %zu\n",
|
|
sizeof(struct in_addr));
|
|
goto drop;
|
|
}
|
|
|
|
if (allow_multicast == 0 && ETHER_IS_MULTICAST(ar_sha(ah))) {
|
|
ARP_LOG(LOG_NOTICE, "%*D is multicast\n",
|
|
ifp->if_addrlen, (u_char *)ar_sha(ah), ":");
|
|
goto drop;
|
|
}
|
|
|
|
op = ntohs(ah->ar_op);
|
|
(void)memcpy(&isaddr, ar_spa(ah), sizeof (isaddr));
|
|
(void)memcpy(&itaddr, ar_tpa(ah), sizeof (itaddr));
|
|
|
|
if (op == ARPOP_REPLY)
|
|
ARPSTAT_INC(rxreplies);
|
|
|
|
/*
|
|
* For a bridge, we want to check the address irrespective
|
|
* of the receive interface. (This will change slightly
|
|
* when we have clusters of interfaces).
|
|
*/
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
|
LIST_FOREACH(ia, INADDR_HASH(itaddr.s_addr), ia_hash) {
|
|
if (((bridged && ia->ia_ifp->if_bridge == ifp->if_bridge) ||
|
|
ia->ia_ifp == ifp) &&
|
|
itaddr.s_addr == ia->ia_addr.sin_addr.s_addr &&
|
|
(ia->ia_ifa.ifa_carp == NULL ||
|
|
(*carp_iamatch_p)(&ia->ia_ifa, &enaddr))) {
|
|
ifa_ref(&ia->ia_ifa);
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
|
goto match;
|
|
}
|
|
}
|
|
LIST_FOREACH(ia, INADDR_HASH(isaddr.s_addr), ia_hash)
|
|
if (((bridged && ia->ia_ifp->if_bridge == ifp->if_bridge) ||
|
|
ia->ia_ifp == ifp) &&
|
|
isaddr.s_addr == ia->ia_addr.sin_addr.s_addr) {
|
|
ifa_ref(&ia->ia_ifa);
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
|
goto match;
|
|
}
|
|
|
|
#define BDG_MEMBER_MATCHES_ARP(addr, ifp, ia) \
|
|
(ia->ia_ifp->if_bridge == ifp->if_softc && \
|
|
!bcmp(IF_LLADDR(ia->ia_ifp), IF_LLADDR(ifp), ifp->if_addrlen) && \
|
|
addr == ia->ia_addr.sin_addr.s_addr)
|
|
/*
|
|
* Check the case when bridge shares its MAC address with
|
|
* some of its children, so packets are claimed by bridge
|
|
* itself (bridge_input() does it first), but they are really
|
|
* meant to be destined to the bridge member.
|
|
*/
|
|
if (is_bridge) {
|
|
LIST_FOREACH(ia, INADDR_HASH(itaddr.s_addr), ia_hash) {
|
|
if (BDG_MEMBER_MATCHES_ARP(itaddr.s_addr, ifp, ia)) {
|
|
ifa_ref(&ia->ia_ifa);
|
|
ifp = ia->ia_ifp;
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
|
goto match;
|
|
}
|
|
}
|
|
}
|
|
#undef BDG_MEMBER_MATCHES_ARP
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
|
|
|
/*
|
|
* No match, use the first inet address on the receive interface
|
|
* as a dummy address for the rest of the function.
|
|
*/
|
|
IF_ADDR_RLOCK(ifp);
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link)
|
|
if (ifa->ifa_addr->sa_family == AF_INET &&
|
|
(ifa->ifa_carp == NULL ||
|
|
(*carp_iamatch_p)(ifa, &enaddr))) {
|
|
ia = ifatoia(ifa);
|
|
ifa_ref(ifa);
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
goto match;
|
|
}
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
|
|
/*
|
|
* If bridging, fall back to using any inet address.
|
|
*/
|
|
IN_IFADDR_RLOCK(&in_ifa_tracker);
|
|
if (!bridged || (ia = TAILQ_FIRST(&V_in_ifaddrhead)) == NULL) {
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
|
goto drop;
|
|
}
|
|
ifa_ref(&ia->ia_ifa);
|
|
IN_IFADDR_RUNLOCK(&in_ifa_tracker);
|
|
match:
|
|
if (!enaddr)
|
|
enaddr = (u_int8_t *)IF_LLADDR(ifp);
|
|
carped = (ia->ia_ifa.ifa_carp != NULL);
|
|
myaddr = ia->ia_addr.sin_addr;
|
|
ifa_free(&ia->ia_ifa);
|
|
if (!bcmp(ar_sha(ah), enaddr, ifp->if_addrlen))
|
|
goto drop; /* it's from me, ignore it. */
|
|
if (!bcmp(ar_sha(ah), ifp->if_broadcastaddr, ifp->if_addrlen)) {
|
|
ARP_LOG(LOG_NOTICE, "link address is broadcast for IP address "
|
|
"%s!\n", inet_ntoa(isaddr));
|
|
goto drop;
|
|
}
|
|
|
|
if (ifp->if_addrlen != ah->ar_hln) {
|
|
ARP_LOG(LOG_WARNING, "from %*D: addr len: new %d, "
|
|
"i/f %d (ignored)\n", ifp->if_addrlen,
|
|
(u_char *) ar_sha(ah), ":", ah->ar_hln,
|
|
ifp->if_addrlen);
|
|
goto drop;
|
|
}
|
|
|
|
/*
|
|
* Warn if another host is using the same IP address, but only if the
|
|
* IP address isn't 0.0.0.0, which is used for DHCP only, in which
|
|
* case we suppress the warning to avoid false positive complaints of
|
|
* potential misconfiguration.
|
|
*/
|
|
if (!bridged && !carped && isaddr.s_addr == myaddr.s_addr &&
|
|
myaddr.s_addr != 0) {
|
|
ARP_LOG(LOG_ERR, "%*D is using my IP address %s on %s!\n",
|
|
ifp->if_addrlen, (u_char *)ar_sha(ah), ":",
|
|
inet_ntoa(isaddr), ifp->if_xname);
|
|
itaddr = myaddr;
|
|
ARPSTAT_INC(dupips);
|
|
goto reply;
|
|
}
|
|
if (ifp->if_flags & IFF_STATICARP)
|
|
goto reply;
|
|
|
|
bzero(&sin, sizeof(sin));
|
|
sin.sin_len = sizeof(struct sockaddr_in);
|
|
sin.sin_family = AF_INET;
|
|
sin.sin_addr = isaddr;
|
|
dst = (struct sockaddr *)&sin;
|
|
IF_AFDATA_RLOCK(ifp);
|
|
la = lla_lookup(LLTABLE(ifp), LLE_EXCLUSIVE, dst);
|
|
IF_AFDATA_RUNLOCK(ifp);
|
|
if (la != NULL)
|
|
arp_check_update_lle(ah, isaddr, ifp, bridged, la);
|
|
else if (itaddr.s_addr == myaddr.s_addr) {
|
|
/*
|
|
* Request/reply to our address, but no lle exists yet.
|
|
* Try to create new llentry.
|
|
*/
|
|
la = lltable_alloc_entry(LLTABLE(ifp), 0, dst);
|
|
if (la == NULL) {
|
|
|
|
/*
|
|
* lle creation may fail if source address belongs
|
|
* to non-directly connected subnet. However, we
|
|
* will try to answer the request instead of dropping
|
|
* frame.
|
|
*/
|
|
goto reply;
|
|
}
|
|
lltable_set_entry_addr(ifp, la, ar_sha(ah));
|
|
|
|
IF_AFDATA_WLOCK(ifp);
|
|
LLE_WLOCK(la);
|
|
la_tmp = lla_lookup(LLTABLE(ifp), LLE_EXCLUSIVE, dst);
|
|
|
|
/*
|
|
* Check if lle still does not exists.
|
|
* If it does, that means that we either
|
|
* 1) have configured it explicitly, via
|
|
* 1a) 'arp -s' static entry or
|
|
* 1b) interface address static record
|
|
* or
|
|
* 2) it was the result of sending first packet to-host
|
|
* or
|
|
* 3) it was another arp reply packet we handled in
|
|
* different thread.
|
|
*
|
|
* In all cases except 3) we definitely need to prefer
|
|
* existing lle. For the sake of simplicity, prefer any
|
|
* existing lle over newly-create one.
|
|
*/
|
|
if (la_tmp == NULL)
|
|
lltable_link_entry(LLTABLE(ifp), la);
|
|
IF_AFDATA_WUNLOCK(ifp);
|
|
|
|
if (la_tmp == NULL) {
|
|
arp_mark_lle_reachable(la);
|
|
LLE_WUNLOCK(la);
|
|
} else {
|
|
/* Free newly-create entry and handle packet */
|
|
lltable_free_entry(LLTABLE(ifp), la);
|
|
la = la_tmp;
|
|
la_tmp = NULL;
|
|
arp_check_update_lle(ah, isaddr, ifp, bridged, la);
|
|
/* arp_check_update_lle() returns @la unlocked */
|
|
}
|
|
la = NULL;
|
|
}
|
|
reply:
|
|
if (op != ARPOP_REQUEST)
|
|
goto drop;
|
|
ARPSTAT_INC(rxrequests);
|
|
|
|
if (itaddr.s_addr == myaddr.s_addr) {
|
|
/* Shortcut.. the receiving interface is the target. */
|
|
(void)memcpy(ar_tha(ah), ar_sha(ah), ah->ar_hln);
|
|
(void)memcpy(ar_sha(ah), enaddr, ah->ar_hln);
|
|
} else {
|
|
struct llentry *lle = NULL;
|
|
|
|
sin.sin_addr = itaddr;
|
|
IF_AFDATA_RLOCK(ifp);
|
|
lle = lla_lookup(LLTABLE(ifp), 0, (struct sockaddr *)&sin);
|
|
IF_AFDATA_RUNLOCK(ifp);
|
|
|
|
if ((lle != NULL) && (lle->la_flags & LLE_PUB)) {
|
|
(void)memcpy(ar_tha(ah), ar_sha(ah), ah->ar_hln);
|
|
(void)memcpy(ar_sha(ah), &lle->ll_addr, ah->ar_hln);
|
|
LLE_RUNLOCK(lle);
|
|
} else {
|
|
|
|
if (lle != NULL)
|
|
LLE_RUNLOCK(lle);
|
|
|
|
if (!V_arp_proxyall)
|
|
goto drop;
|
|
|
|
/* XXX MRT use table 0 for arp reply */
|
|
if (fib4_lookup_nh_basic(0, itaddr, 0, 0, &nh4) != 0)
|
|
goto drop;
|
|
|
|
/*
|
|
* Don't send proxies for nodes on the same interface
|
|
* as this one came out of, or we'll get into a fight
|
|
* over who claims what Ether address.
|
|
*/
|
|
if (nh4.nh_ifp == ifp)
|
|
goto drop;
|
|
|
|
(void)memcpy(ar_tha(ah), ar_sha(ah), ah->ar_hln);
|
|
(void)memcpy(ar_sha(ah), enaddr, ah->ar_hln);
|
|
|
|
/*
|
|
* Also check that the node which sent the ARP packet
|
|
* is on the interface we expect it to be on. This
|
|
* avoids ARP chaos if an interface is connected to the
|
|
* wrong network.
|
|
*/
|
|
|
|
/* XXX MRT use table 0 for arp checks */
|
|
if (fib4_lookup_nh_basic(0, isaddr, 0, 0, &nh4) != 0)
|
|
goto drop;
|
|
if (nh4.nh_ifp != ifp) {
|
|
ARP_LOG(LOG_INFO, "proxy: ignoring request"
|
|
" from %s via %s\n",
|
|
inet_ntoa(isaddr), ifp->if_xname);
|
|
goto drop;
|
|
}
|
|
|
|
#ifdef DEBUG_PROXY
|
|
printf("arp: proxying for %s\n", inet_ntoa(itaddr));
|
|
#endif
|
|
}
|
|
}
|
|
|
|
if (itaddr.s_addr == myaddr.s_addr &&
|
|
IN_LINKLOCAL(ntohl(itaddr.s_addr))) {
|
|
/* RFC 3927 link-local IPv4; always reply by broadcast. */
|
|
#ifdef DEBUG_LINKLOCAL
|
|
printf("arp: sending reply for link-local addr %s\n",
|
|
inet_ntoa(itaddr));
|
|
#endif
|
|
m->m_flags |= M_BCAST;
|
|
m->m_flags &= ~M_MCAST;
|
|
} else {
|
|
/* default behaviour; never reply by broadcast. */
|
|
m->m_flags &= ~(M_BCAST|M_MCAST);
|
|
}
|
|
(void)memcpy(ar_tpa(ah), ar_spa(ah), ah->ar_pln);
|
|
(void)memcpy(ar_spa(ah), &itaddr, ah->ar_pln);
|
|
ah->ar_op = htons(ARPOP_REPLY);
|
|
ah->ar_pro = htons(ETHERTYPE_IP); /* let's be sure! */
|
|
m->m_len = sizeof(*ah) + (2 * ah->ar_pln) + (2 * ah->ar_hln);
|
|
m->m_pkthdr.len = m->m_len;
|
|
m->m_pkthdr.rcvif = NULL;
|
|
sa.sa_family = AF_ARP;
|
|
sa.sa_len = 2;
|
|
m_clrprotoflags(m); /* Avoid confusing lower layers. */
|
|
(*ifp->if_output)(ifp, m, &sa, NULL);
|
|
ARPSTAT_INC(txreplies);
|
|
return;
|
|
|
|
drop:
|
|
m_freem(m);
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Checks received arp data against existing @la.
|
|
* Updates lle state/performs notification if necessary.
|
|
*/
|
|
static void
|
|
arp_check_update_lle(struct arphdr *ah, struct in_addr isaddr, struct ifnet *ifp,
|
|
int bridged, struct llentry *la)
|
|
{
|
|
struct sockaddr sa;
|
|
struct mbuf *m_hold, *m_hold_next;
|
|
|
|
LLE_WLOCK_ASSERT(la);
|
|
|
|
/* the following is not an error when doing bridging */
|
|
if (!bridged && la->lle_tbl->llt_ifp != ifp) {
|
|
if (log_arp_wrong_iface)
|
|
ARP_LOG(LOG_WARNING, "%s is on %s "
|
|
"but got reply from %*D on %s\n",
|
|
inet_ntoa(isaddr),
|
|
la->lle_tbl->llt_ifp->if_xname,
|
|
ifp->if_addrlen, (u_char *)ar_sha(ah), ":",
|
|
ifp->if_xname);
|
|
LLE_WUNLOCK(la);
|
|
return;
|
|
}
|
|
if ((la->la_flags & LLE_VALID) &&
|
|
bcmp(ar_sha(ah), &la->ll_addr, ifp->if_addrlen)) {
|
|
if (la->la_flags & LLE_STATIC) {
|
|
LLE_WUNLOCK(la);
|
|
if (log_arp_permanent_modify)
|
|
ARP_LOG(LOG_ERR,
|
|
"%*D attempts to modify "
|
|
"permanent entry for %s on %s\n",
|
|
ifp->if_addrlen,
|
|
(u_char *)ar_sha(ah), ":",
|
|
inet_ntoa(isaddr), ifp->if_xname);
|
|
return;
|
|
}
|
|
if (log_arp_movements) {
|
|
ARP_LOG(LOG_INFO, "%s moved from %*D "
|
|
"to %*D on %s\n",
|
|
inet_ntoa(isaddr),
|
|
ifp->if_addrlen,
|
|
(u_char *)&la->ll_addr, ":",
|
|
ifp->if_addrlen, (u_char *)ar_sha(ah), ":",
|
|
ifp->if_xname);
|
|
}
|
|
}
|
|
|
|
/* Check if something has changed */
|
|
if (memcmp(&la->ll_addr, ar_sha(ah), ifp->if_addrlen) != 0 ||
|
|
(la->la_flags & LLE_VALID) == 0) {
|
|
/* Perform real LLE update */
|
|
/* use afdata WLOCK to update fields */
|
|
LLE_ADDREF(la);
|
|
LLE_WUNLOCK(la);
|
|
IF_AFDATA_WLOCK(ifp);
|
|
LLE_WLOCK(la);
|
|
|
|
/*
|
|
* Since we droppped LLE lock, other thread might have deleted
|
|
* this lle. Check and return
|
|
*/
|
|
if ((la->la_flags & LLE_DELETED) != 0) {
|
|
IF_AFDATA_WUNLOCK(ifp);
|
|
LLE_FREE_LOCKED(la);
|
|
return;
|
|
}
|
|
|
|
/* Update data */
|
|
lltable_set_entry_addr(ifp, la, ar_sha(ah));
|
|
|
|
IF_AFDATA_WUNLOCK(ifp);
|
|
LLE_REMREF(la);
|
|
|
|
/* Clear fast path feedback request if set */
|
|
la->r_skip_req = 0;
|
|
}
|
|
|
|
arp_mark_lle_reachable(la);
|
|
|
|
/*
|
|
* The packets are all freed within the call to the output
|
|
* routine.
|
|
*
|
|
* NB: The lock MUST be released before the call to the
|
|
* output routine.
|
|
*/
|
|
if (la->la_hold != NULL) {
|
|
m_hold = la->la_hold;
|
|
la->la_hold = NULL;
|
|
la->la_numheld = 0;
|
|
lltable_fill_sa_entry(la, &sa);
|
|
LLE_WUNLOCK(la);
|
|
for (; m_hold != NULL; m_hold = m_hold_next) {
|
|
m_hold_next = m_hold->m_nextpkt;
|
|
m_hold->m_nextpkt = NULL;
|
|
/* Avoid confusing lower layers. */
|
|
m_clrprotoflags(m_hold);
|
|
(*ifp->if_output)(ifp, m_hold, &sa, NULL);
|
|
}
|
|
} else
|
|
LLE_WUNLOCK(la);
|
|
}
|
|
|
|
static void
|
|
arp_mark_lle_reachable(struct llentry *la)
|
|
{
|
|
int canceled, wtime;
|
|
|
|
LLE_WLOCK_ASSERT(la);
|
|
|
|
la->ln_state = ARP_LLINFO_REACHABLE;
|
|
EVENTHANDLER_INVOKE(lle_event, la, LLENTRY_RESOLVED);
|
|
|
|
if (!(la->la_flags & LLE_STATIC)) {
|
|
LLE_ADDREF(la);
|
|
la->la_expire = time_uptime + V_arpt_keep;
|
|
wtime = V_arpt_keep - V_arp_maxtries * V_arpt_rexmit;
|
|
if (wtime < 0)
|
|
wtime = V_arpt_keep;
|
|
canceled = callout_reset(&la->lle_timer,
|
|
hz * wtime, arptimer, la);
|
|
if (canceled)
|
|
LLE_REMREF(la);
|
|
}
|
|
la->la_asked = 0;
|
|
la->la_preempt = V_arp_maxtries;
|
|
}
|
|
|
|
/*
|
|
* Add pernament link-layer record for given interface address.
|
|
*/
|
|
static __noinline void
|
|
arp_add_ifa_lle(struct ifnet *ifp, const struct sockaddr *dst)
|
|
{
|
|
struct llentry *lle, *lle_tmp;
|
|
|
|
/*
|
|
* Interface address LLE record is considered static
|
|
* because kernel code relies on LLE_STATIC flag to check
|
|
* if these entries can be rewriten by arp updates.
|
|
*/
|
|
lle = lltable_alloc_entry(LLTABLE(ifp), LLE_IFADDR | LLE_STATIC, dst);
|
|
if (lle == NULL) {
|
|
log(LOG_INFO, "arp_ifinit: cannot create arp "
|
|
"entry for interface address\n");
|
|
return;
|
|
}
|
|
|
|
IF_AFDATA_WLOCK(ifp);
|
|
LLE_WLOCK(lle);
|
|
/* Unlink any entry if exists */
|
|
lle_tmp = lla_lookup(LLTABLE(ifp), LLE_EXCLUSIVE, dst);
|
|
if (lle_tmp != NULL)
|
|
lltable_unlink_entry(LLTABLE(ifp), lle_tmp);
|
|
|
|
lltable_link_entry(LLTABLE(ifp), lle);
|
|
IF_AFDATA_WUNLOCK(ifp);
|
|
|
|
if (lle_tmp != NULL)
|
|
EVENTHANDLER_INVOKE(lle_event, lle_tmp, LLENTRY_EXPIRED);
|
|
|
|
EVENTHANDLER_INVOKE(lle_event, lle, LLENTRY_RESOLVED);
|
|
LLE_WUNLOCK(lle);
|
|
if (lle_tmp != NULL)
|
|
lltable_free_entry(LLTABLE(ifp), lle_tmp);
|
|
}
|
|
|
|
void
|
|
arp_ifinit(struct ifnet *ifp, struct ifaddr *ifa)
|
|
{
|
|
const struct sockaddr_in *dst_in;
|
|
const struct sockaddr *dst;
|
|
|
|
if (ifa->ifa_carp != NULL)
|
|
return;
|
|
|
|
dst = ifa->ifa_addr;
|
|
dst_in = (const struct sockaddr_in *)dst;
|
|
|
|
if (ntohl(dst_in->sin_addr.s_addr) == INADDR_ANY)
|
|
return;
|
|
arp_announce_addr(ifp, &dst_in->sin_addr, IF_LLADDR(ifp));
|
|
|
|
arp_add_ifa_lle(ifp, dst);
|
|
}
|
|
|
|
void __noinline
|
|
arp_announce_addr(struct ifnet *ifp, const struct in_addr *addr, u_char *enaddr)
|
|
{
|
|
|
|
if (ntohl(addr->s_addr) != INADDR_ANY)
|
|
arprequest(ifp, addr, addr, enaddr);
|
|
}
|
|
|
|
/*
|
|
* Send gratuitous ARPs for all interfaces addresses to notify other nodes of
|
|
* changes.
|
|
*
|
|
* This is a noop if the interface isn't up or has been flagged for no ARP.
|
|
*/
|
|
void __noinline
|
|
arp_announce(struct ifnet *ifp)
|
|
{
|
|
int i, cnt, entries;
|
|
u_char *lladdr;
|
|
struct ifaddr *ifa;
|
|
struct in_addr *addr, *head;
|
|
|
|
if (!(ifp->if_flags & IFF_UP) || (ifp->if_flags & IFF_NOARP))
|
|
return;
|
|
|
|
entries = 8;
|
|
cnt = 0;
|
|
head = malloc(sizeof(*addr) * entries, M_TEMP, M_NOWAIT);
|
|
if (head == NULL) {
|
|
log(LOG_INFO, "arp_announce: malloc %d entries failed\n",
|
|
entries);
|
|
return;
|
|
}
|
|
|
|
/* Take a copy then process to avoid locking issues. */
|
|
IF_ADDR_RLOCK(ifp);
|
|
TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
|
|
if (ifa->ifa_addr->sa_family != AF_INET)
|
|
continue;
|
|
|
|
if (cnt == entries) {
|
|
addr = (struct in_addr *)realloc(head, sizeof(*addr) *
|
|
(entries + 8), M_TEMP, M_NOWAIT);
|
|
if (addr == NULL) {
|
|
log(LOG_INFO, "arp_announce: realloc to %d "
|
|
"entries failed\n", entries + 8);
|
|
/* Process what we have. */
|
|
break;
|
|
}
|
|
entries += 8;
|
|
head = addr;
|
|
}
|
|
|
|
addr = head + cnt;
|
|
bcopy(IFA_IN(ifa), addr, sizeof(*addr));
|
|
cnt++;
|
|
}
|
|
IF_ADDR_RUNLOCK(ifp);
|
|
|
|
lladdr = IF_LLADDR(ifp);
|
|
for (i = 0; i < cnt; i++) {
|
|
arp_announce_addr(ifp, head + i, lladdr);
|
|
}
|
|
free(head, M_TEMP);
|
|
}
|
|
|
|
/*
|
|
* A handler for interface linkstate change events.
|
|
*/
|
|
static void
|
|
arp_ifnet_link_event(void *arg __unused, struct ifnet *ifp, int linkstate)
|
|
{
|
|
|
|
if (linkstate == LINK_STATE_UP && V_arp_on_link)
|
|
arp_announce(ifp);
|
|
}
|
|
|
|
/*
|
|
* A handler for interface link layer address change events.
|
|
*/
|
|
static __noinline void
|
|
arp_iflladdr(void *arg __unused, struct ifnet *ifp)
|
|
{
|
|
|
|
arp_announce(ifp);
|
|
}
|
|
|
|
static void
|
|
arp_init(void)
|
|
{
|
|
|
|
netisr_register(&arp_nh);
|
|
|
|
if (IS_DEFAULT_VNET(curvnet)) {
|
|
iflladdr_tag = EVENTHANDLER_REGISTER(iflladdr_event,
|
|
arp_iflladdr, NULL, EVENTHANDLER_PRI_ANY);
|
|
ifnet_link_event_tag = EVENTHANDLER_REGISTER(ifnet_link_event,
|
|
arp_ifnet_link_event, 0, EVENTHANDLER_PRI_ANY);
|
|
}
|
|
}
|
|
SYSINIT(arp, SI_SUB_PROTO_DOMAIN, SI_ORDER_ANY, arp_init, 0);
|