freebsd-nq/sys/netinet/if_ether.c

1158 lines
32 KiB
C
Raw Normal View History

/*-
1994-05-24 10:09:53 +00:00
* 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"
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/queue.h>
#include <sys/sysctl.h>
1994-05-24 10:09:53 +00:00
#include <sys/systm.h>
#include <sys/mbuf.h>
#include <sys/malloc.h>
2008-09-14 08:24:45 +00:00
#include <sys/proc.h>
#include <sys/rmlock.h>
#include <sys/socket.h>
1994-05-24 10:09:53 +00:00
#include <sys/syslog.h>
#include <net/if.h>
#include <net/if_var.h>
1994-05-24 10:09:53 +00:00
#include <net/if_dl.h>
#include <net/if_types.h>
#include <net/netisr.h>
#include <net/if_llc.h>
#include <net/ethernet.h>
#include <net/route.h>
#include <net/vnet.h>
1994-05-24 10:09:53 +00:00
#include <netinet/in.h>
#include <netinet/in_var.h>
#include <net/if_llatbl.h>
1994-05-24 10:09:53 +00:00
#include <netinet/if_ether.h>
#ifdef INET
#include <netinet/ip_carp.h>
#endif
1994-05-24 10:09:53 +00:00
#include <net/if_arc.h>
#include <net/iso88025.h>
#include <security/mac/mac_framework.h>
#define SIN(s) ((const struct sockaddr_in *)(s))
1994-05-24 10:09:53 +00:00
#define SDL(s) ((struct sockaddr_dl *)s)
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, "");
1994-05-24 10:09:53 +00:00
/* timer values */
static VNET_DEFINE(int, arpt_keep) = (20*60); /* once resolved, good for 20
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator (DPCPU), as suggested by Peter Wemm, and implement a new per-virtual network stack memory allocator. Modify vnet to use the allocator instead of monolithic global container structures (vinet, ...). This change solves many binary compatibility problems associated with VIMAGE, and restores ELF symbols for virtualized global variables. Each virtualized global variable exists as a "reference copy", and also once per virtual network stack. Virtualized global variables are tagged at compile-time, placing the in a special linker set, which is loaded into a contiguous region of kernel memory. Virtualized global variables in the base kernel are linked as normal, but those in modules are copied and relocated to a reserved portion of the kernel's vnet region with the help of a the kernel linker. Virtualized global variables exist in per-vnet memory set up when the network stack instance is created, and are initialized statically from the reference copy. Run-time access occurs via an accessor macro, which converts from the current vnet and requested symbol to a per-vnet address. When "options VIMAGE" is not compiled into the kernel, normal global ELF symbols will be used instead and indirection is avoided. This change restores static initialization for network stack global variables, restores support for non-global symbols and types, eliminates the need for many subsystem constructors, eliminates large per-subsystem structures that caused many binary compatibility issues both for monitoring applications (netstat) and kernel modules, removes the per-function INIT_VNET_*() macros throughout the stack, eliminates the need for vnet_symmap ksym(2) munging, and eliminates duplicate definitions of virtualized globals under VIMAGE_GLOBALS. Bump __FreeBSD_version and update UPDATING. Portions submitted by: bz Reviewed by: bz, zec Discussed with: gnn, jamie, jeff, jhb, julian, sam Suggested by: peter Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
* 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 */
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;
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator (DPCPU), as suggested by Peter Wemm, and implement a new per-virtual network stack memory allocator. Modify vnet to use the allocator instead of monolithic global container structures (vinet, ...). This change solves many binary compatibility problems associated with VIMAGE, and restores ELF symbols for virtualized global variables. Each virtualized global variable exists as a "reference copy", and also once per virtual network stack. Virtualized global variables are tagged at compile-time, placing the in a special linker set, which is loaded into a contiguous region of kernel memory. Virtualized global variables in the base kernel are linked as normal, but those in modules are copied and relocated to a reserved portion of the kernel's vnet region with the help of a the kernel linker. Virtualized global variables exist in per-vnet memory set up when the network stack instance is created, and are initialized statically from the reference copy. Run-time access occurs via an accessor macro, which converts from the current vnet and requested symbol to a per-vnet address. When "options VIMAGE" is not compiled into the kernel, normal global ELF symbols will be used instead and indirection is avoided. This change restores static initialization for network stack global variables, restores support for non-global symbols and types, eliminates the need for many subsystem constructors, eliminates large per-subsystem structures that caused many binary compatibility issues both for monitoring applications (netstat) and kernel modules, removes the per-function INIT_VNET_*() macros throughout the stack, eliminates the need for vnet_symmap ksym(2) munging, and eliminates duplicate definitions of virtualized globals under VIMAGE_GLOBALS. Bump __FreeBSD_version and update UPDATING. Portions submitted by: bz Reviewed by: bz, zec Discussed with: gnn, jamie, jeff, jhb, julian, sam Suggested by: peter Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
#define V_arpt_keep VNET(arpt_keep)
#define V_arpt_down VNET(arpt_down)
#define V_arp_maxtries VNET(arp_maxtries)
#define V_arp_proxyall VNET(arp_proxyall)
#define V_arp_maxhold VNET(arp_maxhold)
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator (DPCPU), as suggested by Peter Wemm, and implement a new per-virtual network stack memory allocator. Modify vnet to use the allocator instead of monolithic global container structures (vinet, ...). This change solves many binary compatibility problems associated with VIMAGE, and restores ELF symbols for virtualized global variables. Each virtualized global variable exists as a "reference copy", and also once per virtual network stack. Virtualized global variables are tagged at compile-time, placing the in a special linker set, which is loaded into a contiguous region of kernel memory. Virtualized global variables in the base kernel are linked as normal, but those in modules are copied and relocated to a reserved portion of the kernel's vnet region with the help of a the kernel linker. Virtualized global variables exist in per-vnet memory set up when the network stack instance is created, and are initialized statically from the reference copy. Run-time access occurs via an accessor macro, which converts from the current vnet and requested symbol to a per-vnet address. When "options VIMAGE" is not compiled into the kernel, normal global ELF symbols will be used instead and indirection is avoided. This change restores static initialization for network stack global variables, restores support for non-global symbols and types, eliminates the need for many subsystem constructors, eliminates large per-subsystem structures that caused many binary compatibility issues both for monitoring applications (netstat) and kernel modules, removes the per-function INIT_VNET_*() macros throughout the stack, eliminates the need for vnet_symmap ksym(2) munging, and eliminates duplicate definitions of virtualized globals under VIMAGE_GLOBALS. Bump __FreeBSD_version and update UPDATING. Portions submitted by: bz Reviewed by: bz, zec Discussed with: gnn, jamie, jeff, jhb, julian, sam Suggested by: peter Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, max_age, CTLFLAG_VNET | CTLFLAG_RW,
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator (DPCPU), as suggested by Peter Wemm, and implement a new per-virtual network stack memory allocator. Modify vnet to use the allocator instead of monolithic global container structures (vinet, ...). This change solves many binary compatibility problems associated with VIMAGE, and restores ELF symbols for virtualized global variables. Each virtualized global variable exists as a "reference copy", and also once per virtual network stack. Virtualized global variables are tagged at compile-time, placing the in a special linker set, which is loaded into a contiguous region of kernel memory. Virtualized global variables in the base kernel are linked as normal, but those in modules are copied and relocated to a reserved portion of the kernel's vnet region with the help of a the kernel linker. Virtualized global variables exist in per-vnet memory set up when the network stack instance is created, and are initialized statically from the reference copy. Run-time access occurs via an accessor macro, which converts from the current vnet and requested symbol to a per-vnet address. When "options VIMAGE" is not compiled into the kernel, normal global ELF symbols will be used instead and indirection is avoided. This change restores static initialization for network stack global variables, restores support for non-global symbols and types, eliminates the need for many subsystem constructors, eliminates large per-subsystem structures that caused many binary compatibility issues both for monitoring applications (netstat) and kernel modules, removes the per-function INIT_VNET_*() macros throughout the stack, eliminates the need for vnet_symmap ksym(2) munging, and eliminates duplicate definitions of virtualized globals under VIMAGE_GLOBALS. Bump __FreeBSD_version and update UPDATING. Portions submitted by: bz Reviewed by: bz, zec Discussed with: gnn, jamie, jeff, jhb, julian, sam Suggested by: peter Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
&VNET_NAME(arpt_keep), 0,
"ARP entry lifetime in seconds");
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, maxtries, CTLFLAG_VNET | CTLFLAG_RW,
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator (DPCPU), as suggested by Peter Wemm, and implement a new per-virtual network stack memory allocator. Modify vnet to use the allocator instead of monolithic global container structures (vinet, ...). This change solves many binary compatibility problems associated with VIMAGE, and restores ELF symbols for virtualized global variables. Each virtualized global variable exists as a "reference copy", and also once per virtual network stack. Virtualized global variables are tagged at compile-time, placing the in a special linker set, which is loaded into a contiguous region of kernel memory. Virtualized global variables in the base kernel are linked as normal, but those in modules are copied and relocated to a reserved portion of the kernel's vnet region with the help of a the kernel linker. Virtualized global variables exist in per-vnet memory set up when the network stack instance is created, and are initialized statically from the reference copy. Run-time access occurs via an accessor macro, which converts from the current vnet and requested symbol to a per-vnet address. When "options VIMAGE" is not compiled into the kernel, normal global ELF symbols will be used instead and indirection is avoided. This change restores static initialization for network stack global variables, restores support for non-global symbols and types, eliminates the need for many subsystem constructors, eliminates large per-subsystem structures that caused many binary compatibility issues both for monitoring applications (netstat) and kernel modules, removes the per-function INIT_VNET_*() macros throughout the stack, eliminates the need for vnet_symmap ksym(2) munging, and eliminates duplicate definitions of virtualized globals under VIMAGE_GLOBALS. Bump __FreeBSD_version and update UPDATING. Portions submitted by: bz Reviewed by: bz, zec Discussed with: gnn, jamie, jeff, jhb, julian, sam Suggested by: peter Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
&VNET_NAME(arp_maxtries), 0,
"ARP resolution attempts before returning error");
SYSCTL_INT(_net_link_ether_inet, OID_AUTO, proxyall, CTLFLAG_VNET | CTLFLAG_RW,
Build on Jeff Roberson's linker-set based dynamic per-CPU allocator (DPCPU), as suggested by Peter Wemm, and implement a new per-virtual network stack memory allocator. Modify vnet to use the allocator instead of monolithic global container structures (vinet, ...). This change solves many binary compatibility problems associated with VIMAGE, and restores ELF symbols for virtualized global variables. Each virtualized global variable exists as a "reference copy", and also once per virtual network stack. Virtualized global variables are tagged at compile-time, placing the in a special linker set, which is loaded into a contiguous region of kernel memory. Virtualized global variables in the base kernel are linked as normal, but those in modules are copied and relocated to a reserved portion of the kernel's vnet region with the help of a the kernel linker. Virtualized global variables exist in per-vnet memory set up when the network stack instance is created, and are initialized statically from the reference copy. Run-time access occurs via an accessor macro, which converts from the current vnet and requested symbol to a per-vnet address. When "options VIMAGE" is not compiled into the kernel, normal global ELF symbols will be used instead and indirection is avoided. This change restores static initialization for network stack global variables, restores support for non-global symbols and types, eliminates the need for many subsystem constructors, eliminates large per-subsystem structures that caused many binary compatibility issues both for monitoring applications (netstat) and kernel modules, removes the per-function INIT_VNET_*() macros throughout the stack, eliminates the need for vnet_symmap ksym(2) munging, and eliminates duplicate definitions of virtualized globals under VIMAGE_GLOBALS. Bump __FreeBSD_version and update UPDATING. Portions submitted by: bz Reviewed by: bz, zec Discussed with: gnn, jamie, jeff, jhb, julian, sam Suggested by: peter Approved by: re (kensmith)
2009-07-14 22:48:30 +00:00
&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");
2002-03-19 21:25:46 +00:00
static void arp_init(void);
static void arpintr(struct mbuf *);
2002-03-19 21:25:46 +00:00
static void arptimer(void *);
#ifdef INET
2002-03-19 21:25:46 +00:00
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_update_lle(struct arphdr *ah, struct ifnet *ifp,
struct llentry *la);
static void arp_mark_lle_reachable(struct llentry *la);
Reimplement the netisr framework in order to support parallel netisr threads: - Support up to one netisr thread per CPU, each processings its own workstream, or set of per-protocol queues. Threads may be bound to specific CPUs, or allowed to migrate, based on a global policy. In the future it would be desirable to support topology-centric policies, such as "one netisr per package". - Allow each protocol to advertise an ordering policy, which can currently be one of: NETISR_POLICY_SOURCE: packets must maintain ordering with respect to an implicit or explicit source (such as an interface or socket). NETISR_POLICY_FLOW: make use of mbuf flow identifiers to place work, as well as allowing protocols to provide a flow generation function for mbufs without flow identifers (m2flow). Falls back on NETISR_POLICY_SOURCE if now flow ID is available. NETISR_POLICY_CPU: allow protocols to inspect and assign a CPU for each packet handled by netisr (m2cpuid). - Provide utility functions for querying the number of workstreams being used, as well as a mapping function from workstream to CPU ID, which protocols may use in work placement decisions. - Add explicit interfaces to get and set per-protocol queue limits, and get and clear drop counters, which query data or apply changes across all workstreams. - Add a more extensible netisr registration interface, in which protocols declare 'struct netisr_handler' structures for each registered NETISR_ type. These include name, handler function, optional mbuf to flow ID function, optional mbuf to CPU ID function, queue limit, and ordering policy. Padding is present to allow these to be expanded in the future. If no queue limit is declared, then a default is used. - Queue limits are now per-workstream, and raised from the previous IFQ_MAXLEN default of 50 to 256. - All protocols are updated to use the new registration interface, and with the exception of netnatm, default queue limits. Most protocols register as NETISR_POLICY_SOURCE, except IPv4 and IPv6, which use NETISR_POLICY_FLOW, and will therefore take advantage of driver- generated flow IDs if present. - Formalize a non-packet based interface between interface polling and the netisr, rather than having polling pretend to be two protocols. Provide two explicit hooks in the netisr worker for start and end events for runs: netisr_poll() and netisr_pollmore(), as well as a function, netisr_sched_poll(), to allow the polling code to schedule netisr execution. DEVICE_POLLING still embeds single-netisr assumptions in its implementation, so for now if it is compiled into the kernel, a single and un-bound netisr thread is enforced regardless of tunable configuration. In the default configuration, the new netisr implementation maintains the same basic assumptions as the previous implementation: a single, un-bound worker thread processes all deferred work, and direct dispatch is enabled by default wherever possible. Performance measurement shows a marginal performance improvement over the old implementation due to the use of batched dequeue. An rmlock is used to synchronize use and registration/unregistration using the framework; currently, synchronized use is disabled (replicating current netisr policy) due to a measurable 3%-6% hit in ping-pong micro-benchmarking. It will be enabled once further rmlock optimization has taken place. However, in practice, netisrs are rarely registered or unregistered at runtime. A new man page for netisr will follow, but since one doesn't currently exist, it hasn't been updated. This change is not appropriate for MFC, although the polling shutdown handler should be merged to 7-STABLE. Bump __FreeBSD_version. Reviewed by: bz
2009-06-01 10:41:38 +00:00
static const struct netisr_handler arp_nh = {
.nh_name = "arp",
.nh_handler = arpintr,
.nh_proto = NETISR_ARP,
.nh_policy = NETISR_POLICY_SOURCE,
};
#ifdef AF_INET
1994-05-24 10:09:53 +00:00
/*
* called by in_scrubprefix() to remove entry from the table when
* the interface goes away
1994-05-24 10:09:53 +00:00
*/
void
arp_ifscrub(struct ifnet *ifp, uint32_t addr)
1994-05-24 10:09:53 +00:00
{
struct sockaddr_in addr4;
bzero((void *)&addr4, sizeof(addr4));
addr4.sin_len = sizeof(addr4);
addr4.sin_family = AF_INET;
addr4.sin_addr.s_addr = addr;
IF_AFDATA_WLOCK(ifp);
lla_delete(LLTABLE(ifp), LLE_IFADDR, (struct sockaddr *)&addr4);
IF_AFDATA_WUNLOCK(ifp);
1994-05-24 10:09:53 +00:00
}
#endif
1994-05-24 10:09:53 +00:00
/*
* Timeout routine. Age arp_tab entries periodically.
1994-05-24 10:09:53 +00:00
*/
static void
arptimer(void *arg)
1994-05-24 10:09:53 +00:00
{
struct llentry *lle = (struct llentry *)arg;
struct ifnet *ifp;
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);
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) {
size_t pkts_dropped;
LLE_REMREF(lle);
pkts_dropped = llentry_free(lle);
ARPSTAT_ADD(dropped, pkts_dropped);
} else
LLE_FREE_LOCKED(lle);
IF_AFDATA_UNLOCK(ifp);
ARPSTAT_INC(timeouts);
CURVNET_RESTORE();
1994-05-24 10:09:53 +00:00
}
/*
* 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)
1994-05-24 10:09:53 +00:00
{
2004-03-14 00:44:11 +00:00
struct mbuf *m;
struct arphdr *ah;
1994-05-24 10:09:53 +00:00
struct sockaddr sa;
u_char *carpaddr = NULL;
1994-05-24 10:09:53 +00:00
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)
1994-05-24 10:09:53 +00:00
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);
1994-05-24 10:09:53 +00:00
}
/*
* 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.
This commit does two things: 1. rt_check() cleanup: rt_check() is only necessary for some address families to gain access to the corresponding arp entry, so call it only in/near the *resolve() routines where it is actually used -- at the moment this is arpresolve(), nd6_storelladdr() (the call is embedded here), and atmresolve() (the call is just before atmresolve to reduce the number of changes). This change will make it a lot easier to decouple the arp table from the routing table. There is an extra call to rt_check() in if_iso88025subr.c to determine the routing info length. I have left it alone for the time being. The interface of arpresolve() and nd6_storelladdr() now changes slightly: + the 'rtentry' parameter (really a hint from the upper level layer) is now passed unchanged from *_output(), so it becomes the route to the final destination and not to the gateway. + the routines will return 0 if resolution is possible, non-zero otherwise. + arpresolve() returns EWOULDBLOCK in case the mbuf is being held waiting for an arp reply -- in this case the error code is masked in the caller so the upper layer protocol will not see a failure. 2. arpcom untangling Where possible, use 'struct ifnet' instead of 'struct arpcom' variables, and use the IFP2AC macro to access arpcom fields. This mostly affects the netatalk code. === Detailed changes: === net/if_arcsubr.c rt_check() cleanup, remove a useless variable net/if_atmsubr.c rt_check() cleanup net/if_ethersubr.c rt_check() cleanup, arpcom untangling net/if_fddisubr.c rt_check() cleanup, arpcom untangling net/if_iso88025subr.c rt_check() cleanup netatalk/aarp.c arpcom untangling, remove a block of duplicated code netatalk/at_extern.h arpcom untangling netinet/if_ether.c rt_check() cleanup (change arpresolve) netinet6/nd6.c rt_check() cleanup (change nd6_storelladdr)
2004-04-25 09:24:52 +00:00
*
* On success, desten and flags are filled in and the function returns 0;
This commit does two things: 1. rt_check() cleanup: rt_check() is only necessary for some address families to gain access to the corresponding arp entry, so call it only in/near the *resolve() routines where it is actually used -- at the moment this is arpresolve(), nd6_storelladdr() (the call is embedded here), and atmresolve() (the call is just before atmresolve to reduce the number of changes). This change will make it a lot easier to decouple the arp table from the routing table. There is an extra call to rt_check() in if_iso88025subr.c to determine the routing info length. I have left it alone for the time being. The interface of arpresolve() and nd6_storelladdr() now changes slightly: + the 'rtentry' parameter (really a hint from the upper level layer) is now passed unchanged from *_output(), so it becomes the route to the final destination and not to the gateway. + the routines will return 0 if resolution is possible, non-zero otherwise. + arpresolve() returns EWOULDBLOCK in case the mbuf is being held waiting for an arp reply -- in this case the error code is masked in the caller so the upper layer protocol will not see a failure. 2. arpcom untangling Where possible, use 'struct ifnet' instead of 'struct arpcom' variables, and use the IFP2AC macro to access arpcom fields. This mostly affects the netatalk code. === Detailed changes: === net/if_arcsubr.c rt_check() cleanup, remove a useless variable net/if_atmsubr.c rt_check() cleanup net/if_ethersubr.c rt_check() cleanup, arpcom untangling net/if_fddisubr.c rt_check() cleanup, arpcom untangling net/if_iso88025subr.c rt_check() cleanup netatalk/aarp.c arpcom untangling, remove a block of duplicated code netatalk/at_extern.h arpcom untangling netinet/if_ether.c rt_check() cleanup (change arpresolve) netinet6/nd6.c rt_check() cleanup (change nd6_storelladdr)
2004-04-25 09:24:52 +00:00
* 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.
1994-05-24 10:09:53 +00:00
*/
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)
1994-05-24 10:09:53 +00:00
{
struct llentry *la = NULL, *la_tmp;
struct mbuf *curr = NULL;
struct mbuf *next = NULL;
int error, renew;
1994-05-24 10:09:53 +00:00
if (pflags != NULL)
*pflags = 0;
if (create == 0) {
IF_AFDATA_RLOCK(ifp);
la = lla_lookup(LLTABLE(ifp), LLE_EXCLUSIVE, dst);
IF_AFDATA_RUNLOCK(ifp);
1994-05-24 10:09:53 +00:00
}
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);
}
2012-08-01 09:00:26 +00:00
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);
2012-08-01 09:00:26 +00:00
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);
renew = 0;
/*
* If entry has an expiry time and it is approaching,
* see if we need to send an ARP request within this
* arpt_down interval.
*/
if (!(la->la_flags & LLE_STATIC) &&
time_uptime + la->la_preempt > la->la_expire) {
renew = 1;
la->la_preempt--;
}
2012-08-01 09:00:26 +00:00
if (pflags != NULL)
*pflags = la->la_flags;
LLE_WUNLOCK(la);
if (renew == 1)
arprequest(ifp, NULL, &SIN(dst)->sin_addr, NULL);
return (0);
}
renew = (la->la_asked == 0 || la->la_expire != time_uptime);
1994-05-24 10:09:53 +00:00
/*
* 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.
1994-05-24 10:09:53 +00:00
*/
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);
1994-05-24 10:09:53 +00:00
}
/*
* 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 lle entry flags.
*
* 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;
int renew;
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), 0, dst);
IF_AFDATA_RUNLOCK(ifp);
if (la == NULL)
return (arpresolve_full(ifp, is_gw, 1, m, dst, desten, pflags));
if ((la->la_flags & LLE_VALID) &&
((la->la_flags & LLE_STATIC) || la->la_expire > time_uptime)) {
bcopy(&la->ll_addr, desten, ifp->if_addrlen);
renew = 0;
/*
* If entry has an expiry time and it is approaching,
* see if we need to send an ARP request within this
* arpt_down interval.
*/
if (!(la->la_flags & LLE_STATIC) &&
time_uptime + la->la_preempt > la->la_expire) {
renew = 1;
la->la_preempt--;
}
if (pflags != NULL)
*pflags = la->la_flags;
LLE_RUNLOCK(la);
if (renew == 1)
arprequest(ifp, NULL, &SIN(dst)->sin_addr, NULL);
return (0);
}
LLE_RUNLOCK(la);
return (arpresolve_full(ifp, is_gw, 0, m, dst, desten, pflags));
}
1994-05-24 10:09:53 +00:00
/*
* Common length and type checks are done here,
* then the protocol-specific routine is called.
*/
static void
arpintr(struct mbuf *m)
1994-05-24 10:09:53 +00:00
{
struct arphdr *ar;
1994-05-24 10:09:53 +00:00
if (m->m_len < sizeof(struct arphdr) &&
((m = m_pullup(m, sizeof(struct arphdr))) == NULL)) {
log(LOG_NOTICE, "arp: runt packet -- m_pullup failed\n");
return;
}
ar = mtod(m, struct arphdr *);
1994-05-24 10:09:53 +00:00
if (ntohs(ar->ar_hrd) != ARPHRD_ETHER &&
ntohs(ar->ar_hrd) != ARPHRD_IEEE802 &&
ntohs(ar->ar_hrd) != ARPHRD_ARCNET &&
ntohs(ar->ar_hrd) != ARPHRD_IEEE1394 &&
ntohs(ar->ar_hrd) != ARPHRD_INFINIBAND) {
log(LOG_NOTICE, "arp: unknown hardware address format (0x%2D)"
" (from %*D to %*D)\n", (unsigned char *)&ar->ar_hrd, "",
ETHER_ADDR_LEN, (u_char *)ar_sha(ar), ":",
ETHER_ADDR_LEN, (u_char *)ar_tha(ar), ":");
m_freem(m);
return;
}
if (m->m_len < arphdr_len(ar)) {
if ((m = m_pullup(m, arphdr_len(ar))) == NULL) {
log(LOG_NOTICE, "arp: runt packet\n");
m_freem(m);
return;
}
ar = mtod(m, struct arphdr *);
}
ARPSTAT_INC(received);
switch (ntohs(ar->ar_pro)) {
#ifdef INET
case ETHERTYPE_IP:
in_arpinput(m);
return;
#endif
1994-05-24 10:09:53 +00:00
}
m_freem(m);
1994-05-24 10:09:53 +00:00
}
#ifdef INET
1994-05-24 10:09:53 +00:00
/*
* 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;
static struct timeval arp_lastlog;
static int arp_curpps;
static int arp_maxpps = 1;
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");
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");
#define ARP_LOG(pri, ...) do { \
if (ppsratecheck(&arp_lastlog, &arp_curpps, arp_maxpps)) \
log((pri), "arp: " __VA_ARGS__); \
} while (0)
1994-05-24 10:09:53 +00:00
static void
in_arpinput(struct mbuf *m)
1994-05-24 10:09:53 +00:00
{
struct rm_priotracker in_ifa_tracker;
2004-03-14 00:44:11 +00:00
struct arphdr *ah;
struct ifnet *ifp = m->m_pkthdr.rcvif;
struct llentry *la = NULL, *la_tmp;
2004-03-14 00:44:11 +00:00
struct rtentry *rt;
struct ifaddr *ifa;
struct in_ifaddr *ia;
1994-05-24 10:09:53 +00:00
struct sockaddr sa;
struct in_addr isaddr, itaddr, myaddr;
u_int8_t *enaddr = NULL;
int op;
int req_len;
int bridged = 0, is_bridge = 0;
int carped;
struct sockaddr_in sin;
struct sockaddr *dst;
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;
req_len = arphdr_len2(ifp->if_addrlen, sizeof(struct in_addr));
if (m->m_len < req_len && (m = m_pullup(m, req_len)) == NULL) {
ARP_LOG(LOG_NOTICE, "runt packet -- m_pullup failed\n");
return;
}
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;
1994-05-24 10:09:53 +00:00
}
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);
1994-05-24 10:09:53 +00:00
itaddr = myaddr;
ARPSTAT_INC(dupips);
1994-05-24 10:09:53 +00:00
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) {
/*
* Reply to our address, but no lle exists yet.
* do we really have to create an entry?
*/
la = lltable_alloc_entry(LLTABLE(ifp), 0, dst);
if (la == NULL)
goto drop;
arp_update_lle(ah, ifp, la);
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;
}
1994-05-24 10:09:53 +00:00
reply:
if (op != ARPOP_REQUEST)
goto drop;
ARPSTAT_INC(rxrequests);
1994-05-24 10:09:53 +00:00
if (itaddr.s_addr == myaddr.s_addr) {
Add code to allow the system to handle multiple routing tables. This particular implementation is designed to be fully backwards compatible and to be MFC-able to 7.x (and 6.x) Currently the only protocol that can make use of the multiple tables is IPv4 Similar functionality exists in OpenBSD and Linux. From my notes: ----- One thing where FreeBSD has been falling behind, and which by chance I have some time to work on is "policy based routing", which allows different packet streams to be routed by more than just the destination address. Constraints: ------------ I want to make some form of this available in the 6.x tree (and by extension 7.x) , but FreeBSD in general needs it so I might as well do it in -current and back port the portions I need. One of the ways that this can be done is to have the ability to instantiate multiple kernel routing tables (which I will now refer to as "Forwarding Information Bases" or "FIBs" for political correctness reasons). Which FIB a particular packet uses to make the next hop decision can be decided by a number of mechanisms. The policies these mechanisms implement are the "Policies" referred to in "Policy based routing". One of the constraints I have if I try to back port this work to 6.x is that it must be implemented as a EXTENSION to the existing ABIs in 6.x so that third party applications do not need to be recompiled in timespan of the branch. This first version will not have some of the bells and whistles that will come with later versions. It will, for example, be limited to 16 tables in the first commit. Implementation method, Compatible version. (part 1) ------------------------------- For this reason I have implemented a "sufficient subset" of a multiple routing table solution in Perforce, and back-ported it to 6.x. (also in Perforce though not always caught up with what I have done in -current/P4). The subset allows a number of FIBs to be defined at compile time (8 is sufficient for my purposes in 6.x) and implements the changes needed to allow IPV4 to use them. I have not done the changes for ipv6 simply because I do not need it, and I do not have enough knowledge of ipv6 (e.g. neighbor discovery) needed to do it. Other protocol families are left untouched and should there be users with proprietary protocol families, they should continue to work and be oblivious to the existence of the extra FIBs. To understand how this is done, one must know that the current FIB code starts everything off with a single dimensional array of pointers to FIB head structures (One per protocol family), each of which in turn points to the trie of routes available to that family. The basic change in the ABI compatible version of the change is to extent that array to be a 2 dimensional array, so that instead of protocol family X looking at rt_tables[X] for the table it needs, it looks at rt_tables[Y][X] when for all protocol families except ipv4 Y is always 0. Code that is unaware of the change always just sees the first row of the table, which of course looks just like the one dimensional array that existed before. The entry points rtrequest(), rtalloc(), rtalloc1(), rtalloc_ign() are all maintained, but refer only to the first row of the array, so that existing callers in proprietary protocols can continue to do the "right thing". Some new entry points are added, for the exclusive use of ipv4 code called in_rtrequest(), in_rtalloc(), in_rtalloc1() and in_rtalloc_ign(), which have an extra argument which refers the code to the correct row. In addition, there are some new entry points (currently called rtalloc_fib() and friends) that check the Address family being looked up and call either rtalloc() (and friends) if the protocol is not IPv4 forcing the action to row 0 or to the appropriate row if it IS IPv4 (and that info is available). These are for calling from code that is not specific to any particular protocol. The way these are implemented would change in the non ABI preserving code to be added later. One feature of the first version of the code is that for ipv4, the interface routes show up automatically on all the FIBs, so that no matter what FIB you select you always have the basic direct attached hosts available to you. (rtinit() does this automatically). You CAN delete an interface route from one FIB should you want to but by default it's there. ARP information is also available in each FIB. It's assumed that the same machine would have the same MAC address, regardless of which FIB you are using to get to it. This brings us as to how the correct FIB is selected for an outgoing IPV4 packet. Firstly, all packets have a FIB associated with them. if nothing has been done to change it, it will be FIB 0. The FIB is changed in the following ways. Packets fall into one of a number of classes. 1/ locally generated packets, coming from a socket/PCB. Such packets select a FIB from a number associated with the socket/PCB. This in turn is inherited from the process, but can be changed by a socket option. The process in turn inherits it on fork. I have written a utility call setfib that acts a bit like nice.. setfib -3 ping target.example.com # will use fib 3 for ping. It is an obvious extension to make it a property of a jail but I have not done so. It can be achieved by combining the setfib and jail commands. 2/ packets received on an interface for forwarding. By default these packets would use table 0, (or possibly a number settable in a sysctl(not yet)). but prior to routing the firewall can inspect them (see below). (possibly in the future you may be able to associate a FIB with packets received on an interface.. An ifconfig arg, but not yet.) 3/ packets inspected by a packet classifier, which can arbitrarily associate a fib with it on a packet by packet basis. A fib assigned to a packet by a packet classifier (such as ipfw) would over-ride a fib associated by a more default source. (such as cases 1 or 2). 4/ a tcp listen socket associated with a fib will generate accept sockets that are associated with that same fib. 5/ Packets generated in response to some other packet (e.g. reset or icmp packets). These should use the FIB associated with the packet being reponded to. 6/ Packets generated during encapsulation. gif, tun and other tunnel interfaces will encapsulate using the FIB that was in effect withthe proces that set up the tunnel. thus setfib 1 ifconfig gif0 [tunnel instructions] will set the fib for the tunnel to use to be fib 1. Routing messages would be associated with their process, and thus select one FIB or another. messages from the kernel would be associated with the fib they refer to and would only be received by a routing socket associated with that fib. (not yet implemented) In addition Netstat has been edited to be able to cope with the fact that the array is now 2 dimensional. (It looks in system memory using libkvm (!)). Old versions of netstat see only the first FIB. In addition two sysctls are added to give: a) the number of FIBs compiled in (active) b) the default FIB of the calling process. Early testing experience: ------------------------- Basically our (IronPort's) appliance does this functionality already using ipfw fwd but that method has some drawbacks. For example, It can't fully simulate a routing table because it can't influence the socket's choice of local address when a connect() is done. Testing during the generating of these changes has been remarkably smooth so far. Multiple tables have co-existed with no notable side effects, and packets have been routes accordingly. ipfw has grown 2 new keywords: setfib N ip from anay to any count ip from any to any fib N In pf there seems to be a requirement to be able to give symbolic names to the fibs but I do not have that capacity. I am not sure if it is required. SCTP has interestingly enough built in support for this, called VRFs in Cisco parlance. it will be interesting to see how that handles it when it suddenly actually does something. Where to next: -------------------- After committing the ABI compatible version and MFCing it, I'd like to proceed in a forward direction in -current. this will result in some roto-tilling in the routing code. Firstly: the current code's idea of having a separate tree per protocol family, all of the same format, and pointed to by the 1 dimensional array is a bit silly. Especially when one considers that there is code that makes assumptions about every protocol having the same internal structures there. Some protocols don't WANT that sort of structure. (for example the whole idea of a netmask is foreign to appletalk). This needs to be made opaque to the external code. My suggested first change is to add routing method pointers to the 'domain' structure, along with information pointing the data. instead of having an array of pointers to uniform structures, there would be an array pointing to the 'domain' structures for each protocol address domain (protocol family), and the methods this reached would be called. The methods would have an argument that gives FIB number, but the protocol would be free to ignore it. When the ABI can be changed it raises the possibilty of the addition of a fib entry into the "struct route". Currently, the structure contains the sockaddr of the desination, and the resulting fib entry. To make this work fully, one could add a fib number so that given an address and a fib, one can find the third element, the fib entry. Interaction with the ARP layer/ LL layer would need to be revisited as well. Qing Li has been working on this already. This work was sponsored by Ironport Systems/Cisco Reviewed by: several including rwatson, bz and mlair (parts each) Obtained from: Ironport systems/Cisco
2008-05-09 23:03:00 +00:00
/* 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);
1994-05-24 10:09:53 +00:00
} 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;
2012-08-01 09:00:26 +00:00
sin.sin_addr = itaddr;
/* XXX MRT use table 0 for arp reply */
rt = in_rtalloc1((struct sockaddr *)&sin, 0, 0UL, 0);
if (!rt)
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 (!rt->rt_ifp || rt->rt_ifp == ifp) {
RTFREE_LOCKED(rt);
goto drop;
}
RTFREE_LOCKED(rt);
(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.
*/
sin.sin_addr = isaddr;
2012-08-01 09:00:26 +00:00
/* XXX MRT use table 0 for arp checks */
rt = in_rtalloc1((struct sockaddr *)&sin, 0, 0UL, 0);
if (!rt)
goto drop;
if (rt->rt_ifp != ifp) {
ARP_LOG(LOG_INFO, "proxy: ignoring request"
" from %s via %s, expecting %s\n",
inet_ntoa(isaddr), ifp->if_xname,
rt->rt_ifp->if_xname);
RTFREE_LOCKED(rt);
goto drop;
}
RTFREE_LOCKED(rt);
#ifdef DEBUG_PROXY
printf("arp: proxying for %s\n", inet_ntoa(itaddr));
#endif
}
1994-05-24 10:09:53 +00:00
}
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);
1994-05-24 10:09:53 +00:00
return;
drop:
m_freem(m);
1994-05-24 10:09:53 +00:00
}
#endif
1994-05-24 10:09:53 +00:00
/*
* 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 */
arp_update_lle(ah, ifp, la);
IF_AFDATA_WUNLOCK(ifp);
LLE_REMREF(la);
}
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);
}
/*
* Updates @la fields used by fast path code.
*/
static void
arp_update_lle(struct arphdr *ah, struct ifnet *ifp, struct llentry *la)
{
memcpy(&la->ll_addr, ar_sha(ah), ifp->if_addrlen);
la->la_flags |= LLE_VALID;
}
static void
arp_mark_lle_reachable(struct llentry *la)
{
int canceled;
LLE_WLOCK_ASSERT(la);
EVENTHANDLER_INVOKE(lle_event, la, LLENTRY_RESOLVED);
if (!(la->la_flags & LLE_STATIC)) {
LLE_ADDREF(la);
la->la_expire = time_uptime + V_arpt_keep;
canceled = callout_reset(&la->lle_timer,
hz * V_arpt_keep, arptimer, la);
if (canceled)
LLE_REMREF(la);
}
la->la_asked = 0;
la->la_preempt = V_arp_maxtries;
}
void
arp_ifinit(struct ifnet *ifp, struct ifaddr *ifa)
{
struct llentry *lle, *lle_tmp;
struct sockaddr_in *dst_in;
struct sockaddr *dst;
if (ifa->ifa_carp != NULL)
return;
ifa->ifa_rtrequest = NULL;
dst_in = IA_SIN(ifa);
dst = (struct sockaddr *)dst_in;
if (ntohl(IA_SIN(ifa)->sin_addr.s_addr) == INADDR_ANY)
return;
arprequest(ifp, &IA_SIN(ifa)->sin_addr,
&IA_SIN(ifa)->sin_addr, IF_LLADDR(ifp));
/*
* 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_ifinit2(struct ifnet *ifp, struct ifaddr *ifa, u_char *enaddr)
{
if (ntohl(IA_SIN(ifa)->sin_addr.s_addr) != INADDR_ANY)
arprequest(ifp, &IA_SIN(ifa)->sin_addr,
&IA_SIN(ifa)->sin_addr, enaddr);
ifa->ifa_rtrequest = NULL;
}
static void
arp_init(void)
{
Reimplement the netisr framework in order to support parallel netisr threads: - Support up to one netisr thread per CPU, each processings its own workstream, or set of per-protocol queues. Threads may be bound to specific CPUs, or allowed to migrate, based on a global policy. In the future it would be desirable to support topology-centric policies, such as "one netisr per package". - Allow each protocol to advertise an ordering policy, which can currently be one of: NETISR_POLICY_SOURCE: packets must maintain ordering with respect to an implicit or explicit source (such as an interface or socket). NETISR_POLICY_FLOW: make use of mbuf flow identifiers to place work, as well as allowing protocols to provide a flow generation function for mbufs without flow identifers (m2flow). Falls back on NETISR_POLICY_SOURCE if now flow ID is available. NETISR_POLICY_CPU: allow protocols to inspect and assign a CPU for each packet handled by netisr (m2cpuid). - Provide utility functions for querying the number of workstreams being used, as well as a mapping function from workstream to CPU ID, which protocols may use in work placement decisions. - Add explicit interfaces to get and set per-protocol queue limits, and get and clear drop counters, which query data or apply changes across all workstreams. - Add a more extensible netisr registration interface, in which protocols declare 'struct netisr_handler' structures for each registered NETISR_ type. These include name, handler function, optional mbuf to flow ID function, optional mbuf to CPU ID function, queue limit, and ordering policy. Padding is present to allow these to be expanded in the future. If no queue limit is declared, then a default is used. - Queue limits are now per-workstream, and raised from the previous IFQ_MAXLEN default of 50 to 256. - All protocols are updated to use the new registration interface, and with the exception of netnatm, default queue limits. Most protocols register as NETISR_POLICY_SOURCE, except IPv4 and IPv6, which use NETISR_POLICY_FLOW, and will therefore take advantage of driver- generated flow IDs if present. - Formalize a non-packet based interface between interface polling and the netisr, rather than having polling pretend to be two protocols. Provide two explicit hooks in the netisr worker for start and end events for runs: netisr_poll() and netisr_pollmore(), as well as a function, netisr_sched_poll(), to allow the polling code to schedule netisr execution. DEVICE_POLLING still embeds single-netisr assumptions in its implementation, so for now if it is compiled into the kernel, a single and un-bound netisr thread is enforced regardless of tunable configuration. In the default configuration, the new netisr implementation maintains the same basic assumptions as the previous implementation: a single, un-bound worker thread processes all deferred work, and direct dispatch is enabled by default wherever possible. Performance measurement shows a marginal performance improvement over the old implementation due to the use of batched dequeue. An rmlock is used to synchronize use and registration/unregistration using the framework; currently, synchronized use is disabled (replicating current netisr policy) due to a measurable 3%-6% hit in ping-pong micro-benchmarking. It will be enabled once further rmlock optimization has taken place. However, in practice, netisrs are rarely registered or unregistered at runtime. A new man page for netisr will follow, but since one doesn't currently exist, it hasn't been updated. This change is not appropriate for MFC, although the polling shutdown handler should be merged to 7-STABLE. Bump __FreeBSD_version. Reviewed by: bz
2009-06-01 10:41:38 +00:00
netisr_register(&arp_nh);
}
SYSINIT(arp, SI_SUB_PROTO_DOMAIN, SI_ORDER_ANY, arp_init, 0);