From nobody@FreeBSD.org  Tue Jan 29 21:48:38 2008
Return-Path: <nobody@FreeBSD.org>
Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34])
	by hub.freebsd.org (Postfix) with ESMTP id 45F5516A419
	for <freebsd-gnats-submit@FreeBSD.org>; Tue, 29 Jan 2008 21:48:38 +0000 (UTC)
	(envelope-from nobody@FreeBSD.org)
Received: from www.freebsd.org (www.freebsd.org [IPv6:2001:4f8:fff6::21])
	by mx1.freebsd.org (Postfix) with ESMTP id 34E1913C45B
	for <freebsd-gnats-submit@FreeBSD.org>; Tue, 29 Jan 2008 21:48:38 +0000 (UTC)
	(envelope-from nobody@FreeBSD.org)
Received: from www.freebsd.org (localhost [127.0.0.1])
	by www.freebsd.org (8.14.2/8.14.2) with ESMTP id m0TLkpg5084751
	for <freebsd-gnats-submit@FreeBSD.org>; Tue, 29 Jan 2008 21:46:51 GMT
	(envelope-from nobody@www.freebsd.org)
Received: (from nobody@localhost)
	by www.freebsd.org (8.14.2/8.14.1/Submit) id m0TLkpBX084750;
	Tue, 29 Jan 2008 21:46:51 GMT
	(envelope-from nobody)
Message-Id: <200801292146.m0TLkpBX084750@www.freebsd.org>
Date: Tue, 29 Jan 2008 21:46:51 GMT
From: Christoph Weber-Fahr <cwf-ml@arcor.de>
Reply-To: cwf-ml@arcor.de
To: freebsd-gnats-submit@FreeBSD.org
Subject: carp causes kernel panics in any constellation
X-Send-Pr-Version: www-3.1
X-GNATS-Notify:

>Number:         120130
>Category:       kern
>Synopsis:       [carp] [panic] carp causes kernel panics in any constellation
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    glebius
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Tue Jan 29 21:50:02 UTC 2008
>Closed-Date:    Wed Jan 22 09:14:14 UTC 2014
>Last-Modified:  Wed Jan 22 09:14:14 UTC 2014
>Originator:     Christoph Weber-Fahr
>Release:        6.3-RELEASE
>Organization:
Arcor AG
>Environment:
FreeBSD XXX.tnd.lab.arcor.de 6.3-RELEASE FreeBSD 6.3-RELEASE #0: Fri Jan 25 21:34:42 CET 2008     wefa@XXX.tnd.lab.arcor.de:/usr/obj/usr/src/sys/DL380  i386

>Description:
Carp reliably and reproducably causes kernel panics.

This is an enhancement of kern/117448 (which itself contains a backreference
to kern/92776). 

The referred PR claims this error only for the case of having and
destroying 2 carp interfaces. We have tested carp extensively, with
both 6.2-RELEASE-p9 and and 6.3-RELEASE, and we have additionally
encountered a number of spontaneous  reboots, spurious lockups and
similar problems.

Note, that even though the reproduction recipe given below is based on
ifconfig destroy commands, we actually saw crashes in normal course of
operation during and between tests where carp was active, both with only
one and with multiple carp interfaces.
>How-To-Repeat:
Currently we also found 2 ways to repeatbly reproduce those effects:

1.) as documented in the referred kern/11744
   ifconfig carp0 destroy
   ifconfig carp1 destroy

This is unrelated to the constellation in which those Interfaces are - in
some constllations the system crashes immediately, in others after the next
ifconfig operation.

2.) is is alsow possible to have a crash using only one crap interface. We
    found the following script to reliably produce a kernel panic within
    15-20 minutes:

   while [ 1 ]
   do
      /etc/rc.d/netif restart
      sleep 35
      ifconfig carp0 destroy
      sleep 35
   done

>Fix:
We do not have a fix.

It should specifically be noted, that using ucarp (from net/ucarp in the
ports collection) is no alternative either. In our tests we found ucarp
1.3 to have serious recovery issues after a failover wich reproducably
left the cluster in a dysfunctional state. We also tested the (not yet
ported) ucarp4 and found it to be completely broken in our environment
(Cisco Switch platform) - they switched the transport to multicast and
apparently completely botched the implementation, so that it doesn't work
on either FreeBSD or Linux.
>Release-Note:
>Audit-Trail:

From: "C. Weber" <cwf-ml@arcor.de>
To: bug-followup@FreeBSD.org
Cc:  
Subject: Re: kern/120130: carp causes kernel panics in any constellation
Date: Tue, 29 Jan 2008 23:33:16 +0100

 Hello,
 
 could someone please correct the From: (Submitter) Address in this
 PR to cwf-ml@arcor.de, or alternatively add a respective Reply-To:
 field with this address.
 
 Thanks,
 
 Christoph Weber-Fahr
 
 [bugmeister note: done.]
Responsible-Changed-From-To: freebsd-bugs->freebsd-net 
Responsible-Changed-By: linimon 
Responsible-Changed-When: Wed Jan 30 00:23:04 UTC 2008 
Responsible-Changed-Why:  
Over to maintainer(s). 

http://www.freebsd.org/cgi/query-pr.cgi?pr=120130 

From: "Max Laier" <max@love2party.net>
To: bug-followup@freebsd.org
Cc: cwf-ml@arcor.de, glebius@freebsd.org
Subject: Re: kern/120130: [carp] [panic] carp causes kernel panics in any 
     constellation
Date: Thu, 31 Jan 2008 22:03:46 +0100 (CET)

 ------=_20080131220346_45398
 Content-Type: text/plain; charset="iso-8859-1"
 Content-Transfer-Encoding: 8bit
 
 Looks like rev. 1.49 of ip_carp.c needs to be MFC'ed.  Patch attached,
 please test and report back.  Thank you.
 
 -- Max
 ------=_20080131220346_45398
 Content-Type: text/x-diff; name="carp.diff"
 Content-Transfer-Encoding: 8bit
 Content-Disposition: attachment; filename="carp.diff"
 
 Index: ip_carp.c
 ===================================================================
 RCS file: /mnt/mlaier/fcvs/src/sys/netinet/ip_carp.c,v
 retrieving revision 1.27.2.11
 diff -u -r1.27.2.11 ip_carp.c
 --- ip_carp.c	6 Jun 2007 16:20:50 -0000	1.27.2.11
 +++ ip_carp.c	31 Jan 2008 21:00:25 -0000
 @@ -1881,8 +1881,12 @@
  				cif = (struct carp_if *)sc->sc_carpdev->if_carp;
  				TAILQ_FOREACH(vr, &cif->vhif_vrs, sc_list)
  					if (vr != sc &&
 -					    vr->sc_vhid == carpr.carpr_vhid)
 -						return EEXIST;
 +					    vr->sc_vhid == carpr.carpr_vhid) {
 +						error = EEXIST;
 +						break;
 +					}
 +				if (error == EEXIST)
 +					break;
  			}
  			sc->sc_vhid = carpr.carpr_vhid;
  			IFP2ENADDR(sc->sc_ifp)[0] = 0;
 ------=_20080131220346_45398--
 

From: Gleb Smirnoff <glebius@FreeBSD.org>
To: Christoph Weber-Fahr <cwf-ml@arcor.de>
Cc: bug-followup@FreeBSD.org
Subject: Re: kern/120130: carp causes kernel panics in any constellation
Date: Fri, 1 Feb 2008 17:04:45 +0300

   Christoph,
 
 > 2.) is is alsow possible to have a crash using only one crap interface. We
 >     found the following script to reliably produce a kernel panic within
 >     15-20 minutes:
 > 
 >    while [ 1 ]
 >    do
 >       /etc/rc.d/netif restart
 >       sleep 35
 >       ifconfig carp0 destroy
 >       sleep 35
 >    done
 
 This reproduce recipe will be complete only if you give configuration
 of network interface in rc.conf. Can you supply this information, please?
 
 Also, have you tried to produce kernel coredump or at least backtrace,
 when kernel panics?
 
 -- 
 Totus tuus, Glebius.
 GLEBIUS-RIPN GLEB-RIPE

From: Christoph Weber-Fahr <wefa2@gmx.de>
To: Gleb Smirnoff <glebius@FreeBSD.org>
Cc: bug-followup@FreeBSD.org
Subject: Re: kern/120130: carp causes kernel panics in any constellation
Date: Sat, 02 Feb 2008 01:33:42 +0100

 Hello,
 
 Gleb Smirnoff <glebius@FreeBSD.org> schrieb am 01.02.2008 15:04:45:
  > This reproduce recipe will be complete only if you give configuration
  > of network interface in rc.conf. Can you supply this information, please?
 
 rc.conf:
 
 ifconfig_em3="inet XX.XX.XX.XX/28 media 100baseTX mediaopt full-duplex"
 firewall_enable="YES"
 firewall_type="open"
 
 cloned_interfaces="carp0"
 ifconfig_carp0="vhid 20 advskew 140 pass heimLich XX.XX.XX.XY netmask 255.255.255.255"
 
 Script:
 
 #!/bin/bash
 while [ 1 ]
 do
 /etc/rc.d/netif restart
 sleep 30
 ifconfig carp0 destroy
 sleep 30
 done
 
 Crash after 10 minutes at 18:25h
 
 I append the first few lines of dmesg at the end.
 
  > Also, have you tried to produce kernel coredump or at least backtrace,
  > when kernel panics?
 
 No. But if you tell me what you want me to do, I could give it a try on
 Monday. Right now I wouldn't even know how to produce kernel dumps and
 backtraces.
 
 Regards
 
 Christoph Weber-Fahr
 
 dmesg:
 
 Copyright (c) 1992-2008 The FreeBSD Project.
 Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
 The Regents of the University of California. All rights reserved.
 FreeBSD is a registered trademark of The FreeBSD Foundation.
 FreeBSD 6.3-RELEASE #0: Fri Jan 25 21:34:42 CET 2008
 root@XX.XX.XX.XX:/usr/obj/usr/src/sys/DL380
 Timecounter "i8254" frequency 1193182 Hz quality 0
 CPU: Intel(R) Xeon(TM) CPU 3.06GHz (3056.51-MHz 686-class CPU)
 Origin = "GenuineIntel"  Id = 0xf29  Stepping = 9
 Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
 Features2=0x4400<CNXT-ID,xTPR>
 Logical CPUs per core: 2
 real memory  = 1073717248 (1023 MB)
 avail memory = 1041592320 (993 MB)
 ACPI APIC Table: <COMPAQ 00000083>
 FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs
 cpu0 (BSP): APIC ID:  6
 cpu1 (AP): APIC ID:  7
 MADT: Forcing active-low polarity and level trigger for SCI
 
Responsible-Changed-From-To: freebsd-net->glebius 
Responsible-Changed-By: glebius 
Responsible-Changed-When: Sat Oct 15 08:19:31 UTC 2011 
Responsible-Changed-Why:  
I'm now working on major rewrite of CARP for FreeBSD 10, and 
I'd like to take all related PRs. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=120130 

From: dfilter@FreeBSD.ORG (dfilter service)
To: bug-followup@FreeBSD.org
Cc:  
Subject: Re: kern/120130: commit references a PR
Date: Fri, 16 Dec 2011 12:17:09 +0000 (UTC)

 Author: glebius
 Date: Fri Dec 16 12:16:56 2011
 New Revision: 228571
 URL: http://svn.freebsd.org/changeset/base/228571
 
 Log:
   A major overhaul of the CARP implementation. The ip_carp.c was started
   from scratch, copying needed functionality from the old implemenation
   on demand, with a thorough review of all code. The main change is that
   interface layer has been removed from the CARP. Now redundant addresses
   are configured exactly on the interfaces, they run on.
   
   The CARP configuration itself is, as before, configured and read via
   SIOCSVH/SIOCGVH ioctls. A new prefix created with SIOCAIFADDR or
   SIOCAIFADDR_IN6 may now be configured to a particular virtual host id,
   which makes the prefix redundant.
   
   ifconfig(8) semantics has been changed too: now one doesn't need
   to clone carpXX interface, he/she should directly configure a vhid
   on a Ethernet interface.
   
   To supply vhid data from the kernel to an application the getifaddrs(8)
   function had been changed to pass ifam_data with each address. [1]
   
   The new implementation definitely closes all PRs related to carp(4)
   being an interface, and may close several others. It also allows
   to run a single redundant IP per interface.
   
   Big thanks to Bjoern Zeeb for his help with inet6 part of patch, for
   idea on using ifam_data and for several rounds of reviewing!
   
   PR:		kern/117000, kern/126945, kern/126714, kern/120130, kern/117448
   Reviewed by:	bz
   Submitted by:	bz [1]
 
 Modified:
   head/UPDATING
   head/lib/libc/net/getifaddrs.c
   head/sbin/ifconfig/af_inet.c
   head/sbin/ifconfig/af_inet6.c
   head/sbin/ifconfig/ifcarp.c
   head/sbin/ifconfig/ifconfig.8
   head/sbin/ifconfig/ifconfig.c
   head/sbin/ifconfig/ifconfig.h
   head/share/man/man4/carp.4
   head/sys/net/if.c
   head/sys/net/if.h
   head/sys/net/if_ethersubr.c
   head/sys/net/if_types.h
   head/sys/net/if_var.h
   head/sys/net/rtsock.c
   head/sys/netinet/if_ether.c
   head/sys/netinet/if_ether.h
   head/sys/netinet/in.c
   head/sys/netinet/in_var.h
   head/sys/netinet/ip_carp.c
   head/sys/netinet/ip_carp.h
   head/sys/netinet6/in6.c
   head/sys/netinet6/in6_ifattach.c
   head/sys/netinet6/in6_var.h
   head/sys/netinet6/nd6.c
   head/sys/netinet6/nd6_nbr.c
   head/sys/sys/param.h
 
 Modified: head/UPDATING
 ==============================================================================
 --- head/UPDATING	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/UPDATING	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -22,6 +22,17 @@ NOTE TO PEOPLE WHO THINK THAT FreeBSD 10
  	machines to maximize performance.  (To disable malloc debugging, run
  	ln -s aj /etc/malloc.conf.)
  
 +20111215:
 +	The carp(4) facility has been changed significantly. Configuration
 +	of the CARP protocol via ifconfig(8) has changed, as well as format
 +	of CARP events submitted to devd(8) has changed. See manual pages
 +	for more information. The arpbalance feature of carp(4) is currently
 +	not supported anymore.
 +
 +	Size of struct in_aliasreq, struct in6_aliasreq has changed. User
 +	utilities using SIOCAIFADDR, SIOCAIFADDR_IN6, e.g. ifconfig(8),
 +	need to be recompiled.
 +
  20111122:
  	The acpi_wmi(4) status device /dev/wmistat has been renamed to
  	/dev/wmistat0.
 
 Modified: head/lib/libc/net/getifaddrs.c
 ==============================================================================
 --- head/lib/libc/net/getifaddrs.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/lib/libc/net/getifaddrs.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -76,7 +76,7 @@ __FBSDID("$FreeBSD$");
  #define	HAVE_IFM_DATA
  #endif
  
 -#if	_BSDI_VERSION >= 199802
 +#if	(_BSDI_VERSION >= 199802) || (__FreeBSD_version >= 1000003) 
  /* ifam_data is very specific to recent versions of bsdi */
  #define	HAVE_IFAM_DATA
  #endif
 
 Modified: head/sbin/ifconfig/af_inet.c
 ==============================================================================
 --- head/sbin/ifconfig/af_inet.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sbin/ifconfig/af_inet.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -84,8 +84,11 @@ in_status(int s __unused, const struct i
  	if (ifa->ifa_flags & IFF_BROADCAST) {
  		sin = (struct sockaddr_in *)ifa->ifa_broadaddr;
  		if (sin != NULL && sin->sin_addr.s_addr != 0)
 -			printf("broadcast %s", inet_ntoa(sin->sin_addr));
 +			printf("broadcast %s ", inet_ntoa(sin->sin_addr));
  	}
 +
 +	print_vhid(ifa, " ");
 +
  	putchar('\n');
  }
  
 
 Modified: head/sbin/ifconfig/af_inet6.c
 ==============================================================================
 --- head/sbin/ifconfig/af_inet6.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sbin/ifconfig/af_inet6.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -307,6 +307,8 @@ in6_status(int s __unused, const struct 
  			printf("infty ");
  	}
  
 +	print_vhid(ifa, " ");
 +
  	putchar('\n');
  }
  
 
 Modified: head/sbin/ifconfig/ifcarp.c
 ==============================================================================
 --- head/sbin/ifconfig/ifcarp.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sbin/ifconfig/ifcarp.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -35,10 +35,11 @@
  #include <stdlib.h>
  #include <unistd.h>
  
 -#include <net/ethernet.h>
  #include <net/if.h>
 +#include <net/if_var.h>
 +#include <netinet/in.h>
 +#include <netinet/in_var.h>
  #include <netinet/ip_carp.h>
 -#include <net/route.h>
  
  #include <ctype.h>
  #include <stdio.h>
 @@ -52,127 +53,153 @@
  
  static const char *carp_states[] = { CARP_STATES };
  
 -void carp_status(int s);
 -void setcarp_advbase(const char *,int, int, const struct afswtch *rafp);
 -void setcarp_advskew(const char *, int, int, const struct afswtch *rafp);
 -void setcarp_passwd(const char *, int, int, const struct afswtch *rafp);
 -void setcarp_vhid(const char *, int, int, const struct afswtch *rafp);
 +static void carp_status(int s);
 +static void setcarp_vhid(const char *, int, int, const struct afswtch *rafp);
 +static void setcarp_callback(int, void *);
 +static void setcarp_advbase(const char *,int, int, const struct afswtch *rafp);
 +static void setcarp_advskew(const char *, int, int, const struct afswtch *rafp);
 +static void setcarp_passwd(const char *, int, int, const struct afswtch *rafp);
 +
 +static int carpr_vhid = -1;
 +static int carpr_advskew = -1;
 +static int carpr_advbase = -1;
 +static int carpr_state = -1;
 +static unsigned char const *carpr_key;
  
 -void
 +static void
  carp_status(int s)
  {
 -	const char *state;
 -	struct carpreq carpr;
 +	struct carpreq carpr[CARP_MAXVHID];
 +	int i;
  
 -	memset((char *)&carpr, 0, sizeof(struct carpreq));
 +	bzero(carpr, sizeof(struct carpreq) * CARP_MAXVHID);
 +	carpr[0].carpr_count = CARP_MAXVHID;
  	ifr.ifr_data = (caddr_t)&carpr;
  
  	if (ioctl(s, SIOCGVH, (caddr_t)&ifr) == -1)
  		return;
  
 -	if (carpr.carpr_vhid > 0) {
 -		if (carpr.carpr_state > CARP_MAXSTATE)
 -			state = "<UNKNOWN>";
 +	for (i = 0; i < carpr[0].carpr_count; i++) {
 +		printf("\tcarp: %s vhid %d advbase %d advskew %d",
 +		    carp_states[carpr[i].carpr_state], carpr[i].carpr_vhid,
 +		    carpr[i].carpr_advbase, carpr[i].carpr_advskew);
 +		if (printkeys && carpr[i].carpr_key[0] != '\0')
 +			printf(" key \"%s\"\n", carpr[i].carpr_key);
  		else
 -			state = carp_states[carpr.carpr_state];
 -
 -		printf("\tcarp: %s vhid %d advbase %d advskew %d\n",
 -		    state, carpr.carpr_vhid, carpr.carpr_advbase,
 -		    carpr.carpr_advskew);
 +			printf("\n");
  	}
 -
 -	return;
 -
  }
  
 -void
 -setcarp_passwd(const char *val, int d, int s, const struct afswtch *afp)
 +static void
 +setcarp_vhid(const char *val, int d, int s, const struct afswtch *afp)
  {
 -	struct carpreq carpr;
  
 -	memset((char *)&carpr, 0, sizeof(struct carpreq));
 -	ifr.ifr_data = (caddr_t)&carpr;
 +	carpr_vhid = atoi(val);
  
 -	if (ioctl(s, SIOCGVH, (caddr_t)&ifr) == -1)
 -		err(1, "SIOCGVH");
 -
 -	memset(carpr.carpr_key, 0, sizeof(carpr.carpr_key));
 -	/* XXX Should hash the password into the key here, perhaps? */
 -	strlcpy(carpr.carpr_key, val, CARP_KEY_LEN);
 -
 -	if (ioctl(s, SIOCSVH, (caddr_t)&ifr) == -1)
 -		err(1, "SIOCSVH");
 +	if (carpr_vhid <= 0 || carpr_vhid > CARP_MAXVHID)
 +		errx(1, "vhid must be greater than 0 and less than %u",
 +		    CARP_MAXVHID);
 +
 +	switch (afp->af_af) {
 +#ifdef INET
 +	case AF_INET:
 +	    {
 +		struct in_aliasreq *ifra;
 +
 +		ifra = (struct in_aliasreq *)afp->af_addreq;
 +		ifra->ifra_vhid = carpr_vhid;
 +		break;
 +	    }
 +#endif
 +#ifdef INET6
 +	case AF_INET6:
 +	    {
 +		struct in6_aliasreq *ifra;
 +
 +		ifra = (struct in6_aliasreq *)afp->af_addreq;
 +		ifra->ifra_vhid = carpr_vhid;
 +		break;
 +	    }
 +#endif
 +	default:
 +		errx(1, "%s doesn't support carp(4)", afp->af_name);
 +	}
  
 -	return;
 +	callback_register(setcarp_callback, NULL);
  }
  
 -void
 -setcarp_vhid(const char *val, int d, int s, const struct afswtch *afp)
 +static void
 +setcarp_callback(int s, void *arg __unused)
  {
 -	int vhid;
  	struct carpreq carpr;
  
 -	vhid = atoi(val);
 -
 -	if (vhid <= 0)
 -		errx(1, "vhid must be greater than 0");
 -
 -	memset((char *)&carpr, 0, sizeof(struct carpreq));
 +	bzero(&carpr, sizeof(struct carpreq));
 +	carpr.carpr_vhid = carpr_vhid;
 +	carpr.carpr_count = 1;
  	ifr.ifr_data = (caddr_t)&carpr;
  
 -	if (ioctl(s, SIOCGVH, (caddr_t)&ifr) == -1)
 +	if (ioctl(s, SIOCGVH, (caddr_t)&ifr) == -1 && errno != ENOENT)
  		err(1, "SIOCGVH");
  
 -	carpr.carpr_vhid = vhid;
 +	if (carpr_key != NULL)
 +		/* XXX Should hash the password into the key here? */
 +		strlcpy(carpr.carpr_key, carpr_key, CARP_KEY_LEN);
 +	if (carpr_advskew > -1)
 +		carpr.carpr_advskew = carpr_advskew;
 +	if (carpr_advbase > -1)
 +		carpr.carpr_advbase = carpr_advbase;
 +	if (carpr_state > -1)
 +		carpr.carpr_state = carpr_state;
  
  	if (ioctl(s, SIOCSVH, (caddr_t)&ifr) == -1)
  		err(1, "SIOCSVH");
 -
 -	return;
  }
  
 -void
 -setcarp_advskew(const char *val, int d, int s, const struct afswtch *afp)
 +static void
 +setcarp_passwd(const char *val, int d, int s, const struct afswtch *afp)
  {
 -	int advskew;
 -	struct carpreq carpr;
  
 -	advskew = atoi(val);
 +	if (carpr_vhid == -1)
 +		errx(1, "passwd requires vhid");
  
 -	memset((char *)&carpr, 0, sizeof(struct carpreq));
 -	ifr.ifr_data = (caddr_t)&carpr;
 +	carpr_key = val;
 +}
  
 -	if (ioctl(s, SIOCGVH, (caddr_t)&ifr) == -1)
 -		err(1, "SIOCGVH");
 +static void
 +setcarp_advskew(const char *val, int d, int s, const struct afswtch *afp)
 +{
  
 -	carpr.carpr_advskew = advskew;
 +	if (carpr_vhid == -1)
 +		errx(1, "advskew requires vhid");
  
 -	if (ioctl(s, SIOCSVH, (caddr_t)&ifr) == -1)
 -		err(1, "SIOCSVH");
 -
 -	return;
 +	carpr_advskew = atoi(val);
  }
  
 -void
 +static void
  setcarp_advbase(const char *val, int d, int s, const struct afswtch *afp)
  {
 -	int advbase;
 -	struct carpreq carpr;
  
 -	advbase = atoi(val);
 +	if (carpr_vhid == -1)
 +		errx(1, "advbase requires vhid");
  
 -	memset((char *)&carpr, 0, sizeof(struct carpreq));
 -	ifr.ifr_data = (caddr_t)&carpr;
 +	carpr_advbase = atoi(val);
 +}
  
 -	if (ioctl(s, SIOCGVH, (caddr_t)&ifr) == -1)
 -		err(1, "SIOCGVH");
 +static void
 +setcarp_state(const char *val, int d, int s, const struct afswtch *afp)
 +{
 +	int i;
  
 -	carpr.carpr_advbase = advbase;
 +	if (carpr_vhid == -1)
 +		errx(1, "state requires vhid");
  
 -	if (ioctl(s, SIOCSVH, (caddr_t)&ifr) == -1)
 -		err(1, "SIOCSVH");
 +	for (i = 0; i <= CARP_MAXSTATE; i++)
 +		if (strcasecmp(carp_states[i], val) == 0) {
 +			carpr_state = i;
 +			return;
 +		}
  
 -	return;
 +	errx(1, "unknown state");
  }
  
  static struct cmd carp_cmds[] = {
 @@ -180,6 +207,7 @@ static struct cmd carp_cmds[] = {
  	DEF_CMD_ARG("advskew",	setcarp_advskew),
  	DEF_CMD_ARG("pass",	setcarp_passwd),
  	DEF_CMD_ARG("vhid",	setcarp_vhid),
 +	DEF_CMD_ARG("state",	setcarp_state),
  };
  static struct afswtch af_carp = {
  	.af_name	= "af_carp",
 
 Modified: head/sbin/ifconfig/ifconfig.8
 ==============================================================================
 --- head/sbin/ifconfig/ifconfig.8	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sbin/ifconfig/ifconfig.8	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -28,7 +28,7 @@
  .\"     From: @(#)ifconfig.8	8.3 (Berkeley) 1/5/94
  .\" $FreeBSD$
  .\"
 -.Dd November 12, 2011
 +.Dd December 16, 2011
  .Dt IFCONFIG 8
  .Os
  .Sh NAME
 @@ -2445,16 +2445,36 @@ The
  argument is useless and hence deprecated.
  .El
  .Pp
 -The following parameters are specific to
 +The following parameters are used to configure
  .Xr carp 4
 -interfaces:
 +protocol on an interface:
  .Bl -tag -width indent
 +.It Cm vhid Ar n
 +Set the virtual host ID.
 +This is a required setting to initiate
 +.Xr carp 4 .
 +If the virtual host ID doesn't exist yet, it is created and attached to the
 +interface, otherwise configuration of an existing vhid is adjusted.
 +If the
 +.Cm vhid
 +keyword is supplied along with an
 +.Dq inet6
 +or
 +.Dq inet
 +address, then this address is configured to be run under control of the
 +specified vhid.
 +Whenever a last address that refers to a particular vhid is removed from an
 +interface, the vhid is automatically removed from interface and destroyed.
 +Any other configuration parameters for the
 +.Xr carp 4
 +protocol should be supplied along with the
 +.Cm vhid
 +keyword.
 +Acceptable values for vhid are 1 to 255.
  .It Cm advbase Ar seconds
  Specifies the base of the advertisement interval in seconds.
  The acceptable values are 1 to 255.
  The default value is 1.
 -.\" The default value is
 -.\" .Dv CARP_DFLTINTV .
  .It Cm advskew Ar interval
  Specifies the skew to add to the base advertisement interval to
  make one host advertise slower than another host.
 @@ -2464,10 +2484,8 @@ The default value is 0.
  .It Cm pass Ar phrase
  Set the authentication key to
  .Ar phrase .
 -.It Cm vhid Ar n
 -Set the virtual host ID.
 -This is a required setting.
 -Acceptable values are 1 to 255.
 +.It Cm state Ar MASTER|BACKUP
 +Forcibly change state of a given vhid.
  .El
  .Pp
  The
 @@ -2530,8 +2548,9 @@ The
  .Fl k
  flag causes keying information for the interface, if available, to be
  printed.
 -For example, the values of 802.11 WEP keys will be printed, if accessible to
 -the current user.
 +For example, the values of 802.11 WEP keys and
 +.Xr carp 4
 +passphrases will be printed, if accessible to the current user.
  This information is not printed by default, as it may be considered
  sensitive.
  .Pp 
 @@ -2593,6 +2612,11 @@ as a synonym for the canonical form of t
  .Fl alias :
  .Dl # ifconfig em0 inet6 2001:db8:bdbd::123/48 delete
  .Pp
 +Configure a single CARP redundant address on igb0, and then switch it
 +to be master:
 +.Dl # ifconfig igb0 vhid 1 10.0.0.1/24 pass foobar
 +.Dl # ifconfig igb0 vhid 1 state master
 +.Pp
  Configure the interface
  .Li xl0 ,
  to use 100baseTX, full duplex Ethernet media options:
 
 Modified: head/sbin/ifconfig/ifconfig.c
 ==============================================================================
 --- head/sbin/ifconfig/ifconfig.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sbin/ifconfig/ifconfig.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -1078,6 +1078,21 @@ printb(const char *s, unsigned v, const 
  }
  
  void
 +print_vhid(const struct ifaddrs *ifa, const char *s)
 +{
 +	struct if_data *ifd;
 +
 +	if (ifa->ifa_data == NULL)
 +		return;
 +
 +	ifd = ifa->ifa_data;
 +	if (ifd->ifi_vhid == 0)
 +		return;
 +	
 +	printf("vhid %d ", ifd->ifi_vhid);
 +}
 +
 +void
  ifmaybeload(const char *name)
  {
  #define MOD_PREFIX_LEN		3	/* "if_" */
 
 Modified: head/sbin/ifconfig/ifconfig.h
 ==============================================================================
 --- head/sbin/ifconfig/ifconfig.h	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sbin/ifconfig/ifconfig.h	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -148,3 +148,6 @@ void	clone_setdefcallback(const char *, 
   * operations on ifmedia can avoid cmd line ordering confusion.
   */
  struct ifmediareq *ifmedia_getstate(int s);
 +
 +void print_vhid(const struct ifaddrs *, const char *);
 +
 
 Modified: head/share/man/man4/carp.4
 ==============================================================================
 --- head/share/man/man4/carp.4	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/share/man/man4/carp.4	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -1,6 +1,7 @@
  .\"	$OpenBSD: carp.4,v 1.16 2004/12/07 23:41:35 jmc Exp $
  .\"
  .\" Copyright (c) 2003, Ryan McBride.  All rights reserved.
 +.\" Copyright (c) 2011, Gleb Smirnoff <glebius@FreeBSD.org>
  .\"
  .\" Redistribution and use in source and binary forms, with or without
  .\" modification, are permitted provided that the following conditions
 @@ -25,7 +26,7 @@
  .\"
  .\" $FreeBSD$
  .\"
 -.Dd August 15, 2011
 +.Dd December 16, 2011
  .Dt CARP 4
  .Os
  .Sh NAME
 @@ -34,33 +35,17 @@
  .Sh SYNOPSIS
  .Cd "device carp"
  .Sh DESCRIPTION
 -The
 -.Nm
 -interface is a pseudo-device that implements and controls the
 -CARP protocol.
 -CARP allows multiple hosts on the same local network to share a set of IP addresses.
 +The CARP allows multiple hosts on the same local network to share a set of
 +IPv4 and/or IPv6 addresses.
  Its primary purpose is to ensure that these
 -addresses are always available, but in some configurations
 -.Nm
 -can also provide load balancing functionality.
 -.Pp
 -A
 -.Nm
 -interface can be created at runtime using the
 -.Nm ifconfig Li carp Ns Ar N Cm create
 -command or by configuring
 -it via
 -.Va cloned_interfaces
 -in the
 -.Pa /etc/rc.conf
 -file.
 +addresses are always available.
  .Pp
  To use
  .Nm ,
 -the administrator needs to configure at minimum a common virtual host ID (VHID)
 -and virtual host IP address on each machine which is to take part in the virtual
 -group.
 -Additional parameters can also be set on a per-interface basis:
 +the administrator needs to configure at minimum a common virtual host ID
 +(vhid) and attach at least one IP address to this vhid on each machine which
 +is to take part in the virtual group.
 +Additional parameters can also be set on a per-vhid basis:
  .Cm advbase
  and
  .Cm advskew ,
 @@ -93,9 +78,20 @@ or through the
  .Dv SIOCSVH
  .Xr ioctl 2 .
  .Pp
 +CARP virtual hosts can be configured on multicast capable interfaces: Ethernet,
 +layer 2 VLAN, FDDI and Token Ring.
 +An arbitrary number of virtual host IDs can be configured on an interface.
 +An arbitrary number of IPv4 or IPv6 addresses can be attached to a particular
 +vhid.
 +It is important that all hosts participating in a vhid have the same list
 +of prefixes configured on the vhid, since all prefixes are included in the
 +cryptographic checksum supplied in each advertisement.
 +Multiple vhids running on one interface participate in master/backup
 +elections independently.
 +.Pp
  Additionally, there are a number of global parameters which can be set using
  .Xr sysctl 8 :
 -.Bl -tag -width ".Va net.inet.carp.arpbalance"
 +.Bl -tag -width ".Va net.inet.carp.preempt"
  .It Va net.inet.carp.allow
  Accept incoming
  .Nm
 @@ -125,9 +121,6 @@ Values above 1 enable logging of bad
  .Nm
  packets.
  Default value is 1.
 -.It Va net.inet.carp.arpbalance
 -Balance local traffic using ARP (see below).
 -Disabled by default.
  .It Va net.inet.carp.suppress_preempt
  A read only value showing the status of preemption suppression.
  Preemption can be suppressed if link on an interface is down
 @@ -138,36 +131,36 @@ Value of 0 means that preemption is not 
  problems are detected.
  Every problem increments suppression counter.
  .El
 -.Sh ARP level load balancing
 -The
 -.Nm
 -has limited abilities for load balancing the incoming connections
 -between hosts in Ethernet network.
 -For load balancing operation, one needs several CARP interfaces that
 -are configured to the same IP address, but to a different VHIDs.
 -Once an ARP request is received, the CARP protocol will use a hashing
 -function against the source IP address in the ARP request to determine
 -which VHID should this request belong to.
 -If the corresponding CARP interface is in master state, the ARP request
 -will be replied, otherwise it will be ignored.
 -See the
 -.Sx EXAMPLES
 -section for a practical example of load balancing.
 -.Pp
 -The ARP load balancing has some limitations.
 -First, ARP balancing only works on the local network segment.
 -It cannot balance traffic that crosses a router, because the
 -router itself will always be balanced to the same virtual host.
 -Second, ARP load balancing can lead to asymmetric routing
 -of incoming and outgoing traffic, and thus combining it with
 -.Xr pfsync 4
 -is dangerous, because this creates a race condition between
 -balanced routers and a host they are serving.
 -Imagine an incoming packet creating state on the first router, being
 -forwarded to its destination, and destination replying faster
 -than the state information is packed and synced with the second router.
 -If the reply would be load balanced to second router, it will be
 -dropped due to no state.
 +.\".Sh ARP level load balancing
 +.\"The
 +.\".Nm
 +.\"has limited abilities for load balancing the incoming connections
 +.\"between hosts in Ethernet network.
 +.\"For load balancing operation, one needs several CARP interfaces that
 +.\"are configured to the same IP address, but to a different vhids.
 +.\"Once an ARP request is received, the CARP protocol will use a hashing
 +.\"function against the source IP address in the ARP request to determine
 +.\"which vhid should this request belong to.
 +.\"If the corresponding CARP interface is in master state, the ARP request
 +.\"will be replied, otherwise it will be ignored.
 +.\"See the
 +.\".Sx EXAMPLES
 +.\"section for a practical example of load balancing.
 +.\".Pp
 +.\"The ARP load balancing has some limitations.
 +.\"First, ARP balancing only works on the local network segment.
 +.\"It cannot balance traffic that crosses a router, because the
 +.\"router itself will always be balanced to the same virtual host.
 +.\"Second, ARP load balancing can lead to asymmetric routing
 +.\"of incoming and outgoing traffic, and thus combining it with
 +.\".Xr pfsync 4
 +.\"is dangerous, because this creates a race condition between
 +.\"balanced routers and a host they are serving.
 +.\"Imagine an incoming packet creating state on the first router, being
 +.\"forwarded to its destination, and destination replying faster
 +.\"than the state information is packed and synced with the second router.
 +.\"If the reply would be load balanced to second router, it will be
 +.\"dropped due to no state.
  .Sh STATE CHANGE NOTIFICATIONS
  Sometimes it is useful to get notified about
  .Nm
 @@ -175,13 +168,10 @@ status change events.
  This can be accomplished by using
  .Xr devd 8
  hooks.
 -Master/slave events are signalled as
 -.Nm
 -interface
 -.Dv LINK_UP
 -or
 -.Dv LINK_DOWN
 -event.
 +Master/slave events are signalled under system
 +.Dv CARP .
 +Subsystem specifies vhid and name of interface, where event occured.
 +Type of the message displays new state of vhid.
  Please see
  .Xr devd.conf 5
  and 
 @@ -197,23 +187,19 @@ Enable it on both host A and B:
  .Pp
  .Dl sysctl net.inet.carp.preempt=1
  .Pp
 -Assume that host A is the preferred master and 192.168.1.x/24 is
 -configured on one physical interface and 192.168.2.y/24 on another.
 +Assume that host A is the preferred master and we are running the
 +192.168.1.0/24 prefix on em0 and 192.168.2.0/24 on em1.
  This is the setup for host A:
  .Bd -literal -offset indent
 -ifconfig carp0 create
 -ifconfig carp0 vhid 1 pass mekmitasdigoat 192.168.1.1/24
 -ifconfig carp1 create
 -ifconfig carp1 vhid 2 pass mekmitasdigoat 192.168.2.1/24
 +ifconfig em0 vhid 1 pass mekmitasdigoat 192.168.1.1/24
 +ifconfig em1 vhid 2 pass mekmitasdigoat 192.168.2.1/24
  .Ed
  .Pp
  The setup for host B is identical, but it has a higher
  .Cm advskew :
  .Bd -literal -offset indent
 -ifconfig carp0 create
 -ifconfig carp0 vhid 1 advskew 100 pass mekmitasdigoat 192.168.1.1/24
 -ifconfig carp1 create
 -ifconfig carp1 vhid 2 advskew 100 pass mekmitasdigoat 192.168.2.1/24
 +ifconfig em0 vhid 1 advskew 100 pass mekmitasdigoat 192.168.1.1/24
 +ifconfig em1 vhid 2 advskew 100 pass mekmitasdigoat 192.168.2.1/24
  .Ed
  .Pp
  Because of the preempt option, when one of the physical interfaces of
 @@ -224,67 +210,60 @@ is adjusted to 240 on all its
  interfaces.
  This will cause host B to preempt on both interfaces instead of
  just the failed one.
 -.Pp
 -In order to set up an ARP balanced virtual host, it is necessary to configure
 -one virtual host for each physical host which would respond to ARP requests
 -and thus handle the traffic.
 -In the following example, two virtual hosts are configured on two hosts to
 -provide balancing and failover for the IP address 192.168.1.10.
 -.Pp
 -First the
 -.Nm
 -interfaces on host A are configured.
 -The
 -.Cm advskew
 -of 100 on the second virtual host means that its advertisements will be sent
 -out slightly less frequently.
 -.Bd -literal -offset indent
 -ifconfig carp0 create
 -ifconfig carp0 vhid 1 pass mekmitasdigoat 192.168.1.10/24
 -ifconfig carp1 create
 -ifconfig carp1 vhid 2 advskew 100 pass mekmitasdigoat 192.168.1.10/24
 -.Ed
 -.Pp
 -The configuration for host B is identical, except the
 -.Cm advskew
 -is on virtual host 1 rather than virtual host 2.
 -.Bd -literal -offset indent
 -ifconfig carp0 create
 -ifconfig carp0 vhid 1 advskew 100 pass mekmitasdigoat 192.168.1.10/24
 -ifconfig carp1 create
 -ifconfig carp1 vhid 2 pass mekmitasdigoat 192.168.1.10/24
 -.Ed
 -.Pp
 -Finally, the ARP balancing feature must be enabled on both hosts:
 -.Pp
 -.Dl sysctl net.inet.carp.arpbalance=1
 -.Pp
 -When the hosts receive an ARP request for 192.168.1.10, the source IP address
 -of the request is used to compute which virtual host should answer the request.
 -The host which is master of the selected virtual host will reply to the
 -request, the other(s) will ignore it.
 -.Pp
 -This way, locally connected systems will receive different ARP replies and
 -subsequent IP traffic will be balanced among the hosts.
 -If one of the hosts fails, the other will take over the virtual MAC address,
 -and begin answering ARP requests on its behalf.
 +.\".Pp
 +.\"In order to set up an ARP balanced virtual host, it is necessary to configure
 +.\"one virtual host for each physical host which would respond to ARP requests
 +.\"and thus handle the traffic.
 +.\"In the following example, two virtual hosts are configured on two hosts to
 +.\"provide balancing and failover for the IP address 192.168.1.10.
 +.\".Pp
 +.\"First the
 +.\".Nm
 +.\"interfaces on host A are configured.
 +.\"The
 +.\".Cm advskew
 +.\"of 100 on the second virtual host means that its advertisements will be sent
 +.\"out slightly less frequently.
 +.\".Bd -literal -offset indent
 +.\"ifconfig carp0 create
 +.\"ifconfig carp0 vhid 1 pass mekmitasdigoat 192.168.1.10/24
 +.\"ifconfig carp1 create
 +.\"ifconfig carp1 vhid 2 advskew 100 pass mekmitasdigoat 192.168.1.10/24
 +.\".Ed
 +.\".Pp
 +.\"The configuration for host B is identical, except the
 +.\".Cm advskew
 +.\"is on virtual host 1 rather than virtual host 2.
 +.\".Bd -literal -offset indent
 +.\"ifconfig carp0 create
 +.\"ifconfig carp0 vhid 1 advskew 100 pass mekmitasdigoat 192.168.1.10/24
 +.\"ifconfig carp1 create
 +.\"ifconfig carp1 vhid 2 pass mekmitasdigoat 192.168.1.10/24
 +.\".Ed
 +.\".Pp
 +.\"Finally, the ARP balancing feature must be enabled on both hosts:
 +.\".Pp
 +.\".Dl sysctl net.inet.carp.arpbalance=1
 +.\".Pp
 +.\"When the hosts receive an ARP request for 192.168.1.10, the source IP address
 +.\"of the request is used to compute which virtual host should answer the request.
 +.\"The host which is master of the selected virtual host will reply to the
 +.\"request, the other(s) will ignore it.
 +.\".Pp
 +.\"This way, locally connected systems will receive different ARP replies and
 +.\"subsequent IP traffic will be balanced among the hosts.
 +.\"If one of the hosts fails, the other will take over the virtual MAC address,
 +.\"and begin answering ARP requests on its behalf.
  .Pp
  Processing of
  .Nm
 -status change events can be set up by using the following devd.conf rules:
 +status change events can be set up by using the following devd.conf rule:
  .Bd -literal -offset indent
  notify 0 {
 -	match "system"          "IFNET";
 -	match "type"            "LINK_UP";
 -	match "subsystem"       "carp*";
 -	action "/root/carpcontrol.sh $type $subsystem";
 -};
 -
 -notify 0 {
 -	match "system"          "IFNET";
 -	match "type"            "LINK_DOWN";
 -	match "subsystem"       "carp*";
 -	action "/root/carpcontrol.sh $type $subsystem";
 +	match "system"          "CARP";
 +	match "subsystem"       "[0-9]+@";
 +	match "type"            "(MASTER|BACKUP)";
 +	action "/root/carpcontrol.sh $subsystem $type";
  };
  .Ed
  .Sh SEE ALSO
 @@ -303,3 +282,8 @@ The
  .Nm
  device was imported into
  .Fx 5.4 .
 +In
 +.Fx 10
 +the
 +.Nm
 +was significantly rewritten, and is no longer a pseudo-interface.
 
 Modified: head/sys/net/if.c
 ==============================================================================
 --- head/sys/net/if.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/net/if.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -130,17 +130,19 @@ void	(*lagg_linkstate_p)(struct ifnet *i
  /* These are external hooks for CARP. */
  void	(*carp_linkstate_p)(struct ifnet *ifp);
  #if defined(INET) || defined(INET6)
 -struct ifnet *(*carp_forus_p)(struct ifnet *ifp, u_char *dhost);
 +int	(*carp_forus_p)(struct ifnet *ifp, u_char *dhost);
  int	(*carp_output_p)(struct ifnet *ifp, struct mbuf *m,
 -    struct sockaddr *sa, struct rtentry *rt);
 +    struct sockaddr *sa);
 +int	(*carp_ioctl_p)(struct ifreq *, u_long, struct thread *);   
 +int	(*carp_attach_p)(struct ifaddr *, int);
 +void	(*carp_detach_p)(struct ifaddr *);
  #endif
  #ifdef INET
 -int (*carp_iamatch_p)(struct ifnet *, struct in_ifaddr *, struct in_addr *,
 -    u_int8_t **);
 +int	(*carp_iamatch_p)(struct ifaddr *, uint8_t **);
  #endif
  #ifdef INET6
  struct ifaddr *(*carp_iamatch6_p)(struct ifnet *ifp, struct in6_addr *taddr6);
 -caddr_t (*carp_macmatch6_p)(struct ifnet *ifp, struct mbuf *m,
 +caddr_t	(*carp_macmatch6_p)(struct ifnet *ifp, struct mbuf *m,
      const struct in6_addr *taddr);
  #endif
  
 @@ -2506,6 +2508,16 @@ ifioctl(struct socket *so, u_long cmd, c
  		error = if_getgroupmembers((struct ifgroupreq *)data);
  		CURVNET_RESTORE();
  		return (error);
 +#if defined(INET) || defined(INET6)
 +	case SIOCSVH:
 +	case SIOCGVH:
 +		if (carp_ioctl_p == NULL)
 +			error = EPROTONOSUPPORT;
 +		else
 +			error = (*carp_ioctl_p)(ifr, cmd, td);
 +		CURVNET_RESTORE();
 +		return (error);
 +#endif
  	}
  
  	ifp = ifunit_ref(ifr->ifr_name);
 
 Modified: head/sys/net/if.h
 ==============================================================================
 --- head/sys/net/if.h	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/net/if.h	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -85,7 +85,7 @@ struct if_data {
  	u_char	ifi_addrlen;		/* media address length */
  	u_char	ifi_hdrlen;		/* media header length */
  	u_char	ifi_link_state;		/* current link state */
 -	u_char	ifi_spare_char1;	/* spare byte */
 +	u_char	ifi_vhid;		/* carp vhid */
  	u_char	ifi_spare_char2;	/* spare byte */
  	u_char	ifi_datalen;		/* length of this data struct */
  	u_long	ifi_mtu;		/* maximum transmission unit */
 @@ -267,6 +267,8 @@ struct ifa_msghdr {
  	int	ifam_flags;	/* value of ifa_flags */
  	u_short	ifam_index;	/* index for associated ifp */
  	int	ifam_metric;	/* value of ifa_metric */
 +	struct	if_data ifam_data;/* statistics and other data about if or
 +				 * address */
  };
  
  /*
 @@ -357,6 +359,7 @@ struct ifaliasreq {
  	struct	sockaddr ifra_addr;
  	struct	sockaddr ifra_broadaddr;
  	struct	sockaddr ifra_mask;
 +	int	ifra_vhid;
  };
  
  struct ifmediareq {
 
 Modified: head/sys/net/if_ethersubr.c
 ==============================================================================
 --- head/sys/net/if_ethersubr.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/net/if_ethersubr.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -397,7 +397,7 @@ ether_output(struct ifnet *ifp, struct m
  
  #if defined(INET) || defined(INET6)
  	if (ifp->if_carp &&
 -	    (error = (*carp_output_p)(ifp, m, dst, NULL)))
 +	    (error = (*carp_output_p)(ifp, m, dst)))
  		goto bad;
  #endif
  
 
 Modified: head/sys/net/if_types.h
 ==============================================================================
 --- head/sys/net/if_types.h	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/net/if_types.h	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -250,6 +250,5 @@
  #define	IFT_ENC		0xf4
  #define	IFT_PFLOG	0xf6
  #define	IFT_PFSYNC	0xf7
 -#define	IFT_CARP	0xf8	/* Common Address Redundancy Protocol */
  #define IFT_IPXIP	0xf9	/* IPX over IP tunneling; no longer used. */
  #endif /* !_NET_IF_TYPES_H_ */
 
 Modified: head/sys/net/if_var.h
 ==============================================================================
 --- head/sys/net/if_var.h	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/net/if_var.h	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -69,6 +69,7 @@ struct	rt_addrinfo;
  struct	socket;
  struct	ether_header;
  struct	carp_if;
 +struct	carp_softc;
  struct  ifvlantrunk;
  struct	route;
  struct	vnet;
 @@ -729,6 +730,7 @@ struct ifaddr {
  	struct	sockaddr *ifa_netmask;	/* used to determine subnet */
  	struct	if_data if_data;	/* not all members are meaningful */
  	struct	ifnet *ifa_ifp;		/* back-pointer to interface */
 +	struct	carp_softc *ifa_carp;	/* pointer to CARP data */
  	TAILQ_ENTRY(ifaddr) ifa_link;	/* queue macro glue */
  	void	(*ifa_rtrequest)	/* check or clean routes (+ or -)'d */
  		(int, struct rtentry *, struct rt_addrinfo *);
 
 Modified: head/sys/net/rtsock.c
 ==============================================================================
 --- head/sys/net/rtsock.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/net/rtsock.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -63,6 +63,7 @@
  
  #include <netinet/in.h>
  #include <netinet/if_ether.h>
 +#include <netinet/ip_carp.h>
  #ifdef INET6
  #include <netinet6/scope6_var.h>
  #endif
 @@ -83,7 +84,7 @@ struct if_data32 {
  	uint8_t	ifi_addrlen;
  	uint8_t	ifi_hdrlen;
  	uint8_t	ifi_link_state;
 -	uint8_t	ifi_spare_char1;
 +	uint8_t	ifi_vhid;
  	uint8_t	ifi_spare_char2;
  	uint8_t	ifi_datalen;
  	uint32_t ifi_mtu;
 @@ -122,6 +123,9 @@ MALLOC_DEFINE(M_RTABLE, "routetbl", "rou
  static struct	sockaddr route_src = { 2, PF_ROUTE, };
  static struct	sockaddr sa_zero   = { sizeof(sa_zero), AF_INET, };
  
 +/* These are external hooks for CARP. */
 +int	(*carp_get_vhid_p)(struct ifaddr *);
 +
  /*
   * Used by rtsock/raw_input callback code to decide whether to filter the update
   * notification to a socket bound to a particular FIB.
 @@ -1508,6 +1512,7 @@ copy_ifdata32(struct if_data *src, struc
  	CP(*src, *dst, ifi_addrlen);
  	CP(*src, *dst, ifi_hdrlen);
  	CP(*src, *dst, ifi_link_state);
 +	CP(*src, *dst, ifi_vhid);
  	dst->ifi_datalen = sizeof(struct if_data32);
  	CP(*src, *dst, ifi_mtu);
  	CP(*src, *dst, ifi_metric);
 @@ -1559,6 +1564,9 @@ sysctl_iflist(int af, struct walkarg *w)
  				ifm32->ifm_flags = ifp->if_flags |
  				    ifp->if_drv_flags;
  				copy_ifdata32(&ifp->if_data, &ifm32->ifm_data);
 +				if (carp_get_vhid_p != NULL)
 +					ifm32->ifm_data.ifi_vhid =
 +					    (*carp_get_vhid_p)(ifa);
  				ifm32->ifm_addrs = info.rti_addrs;
  				error = SYSCTL_OUT(w->w_req, (caddr_t)ifm32,
  				    len);
 @@ -1569,6 +1577,9 @@ sysctl_iflist(int af, struct walkarg *w)
  			ifm->ifm_index = ifp->if_index;
  			ifm->ifm_flags = ifp->if_flags | ifp->if_drv_flags;
  			ifm->ifm_data = ifp->if_data;
 +			if (carp_get_vhid_p != NULL)
 +				ifm->ifm_data.ifi_vhid =
 +				    (*carp_get_vhid_p)(ifa);
  			ifm->ifm_addrs = info.rti_addrs;
  			error = SYSCTL_OUT(w->w_req, (caddr_t)ifm, len);
  #ifdef COMPAT_FREEBSD32
 @@ -1595,6 +1606,9 @@ sysctl_iflist(int af, struct walkarg *w)
  				ifam->ifam_flags = ifa->ifa_flags;
  				ifam->ifam_metric = ifa->ifa_metric;
  				ifam->ifam_addrs = info.rti_addrs;
 +				if (carp_get_vhid_p != NULL)
 +					ifam->ifam_data.ifi_vhid =
 +					    (*carp_get_vhid_p)(ifa);
  				error = SYSCTL_OUT(w->w_req, w->w_tmem, len);
  				if (error)
  					goto done;
 
 Modified: head/sys/netinet/if_ether.c
 ==============================================================================
 --- head/sys/netinet/if_ether.c	Fri Dec 16 11:52:33 2011	(r228570)
 +++ head/sys/netinet/if_ether.c	Fri Dec 16 12:16:56 2011	(r228571)
 @@ -139,8 +139,6 @@ static const struct netisr_handler arp_n
  };
  
  #ifdef AF_INET
 -void arp_ifscrub(struct ifnet *ifp, uint32_t addr);
 -
  /*
   * called by in_ifscrub to remove entry from the table when
   * the interface goes away
 @@ -516,7 +514,7 @@ in_arpinput(struct mbuf *m)
  	int op, flags;
  	int req_len;
  	int bridged = 0, is_bridge = 0;
 -	int carp_match = 0;
 +	int carped;
  	struct sockaddr_in sin;
  	sin.sin_len = sizeof(struct sockaddr_in);
  	sin.sin_family = AF_INET;
 @@ -561,24 +559,14 @@ in_arpinput(struct mbuf *m)
  	 * For a bridge, we want to check the address irrespective
  	 * of the receive interface. (This will change slightly
  	 * when we have clusters of interfaces).
 -	 * If the interface does not match, but the recieving interface
 -	 * is part of carp, we call carp_iamatch to see if this is a
 -	 * request for the virtual host ip.
 -	 * XXX: This is really ugly!
  	 */
  	IN_IFADDR_RLOCK();
  	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) {
 -			ifa_ref(&ia->ia_ifa);
 -			IN_IFADDR_RUNLOCK();
 -			goto match;
 -		}
 -		if (ifp->if_carp != NULL &&
 -		    (*carp_iamatch_p)(ifp, ia, &isaddr, &enaddr) &&
 -		    itaddr.s_addr == ia->ia_addr.sin_addr.s_addr) {
 -			carp_match = 1;
 +		    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();
  			goto match;
 @@ -643,6 +631,7 @@ in_arpinput(struct mbuf *m)
  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))
 @@ -659,9 +648,9 @@ match:
  	 * case we suppress the warning to avoid false positive complaints of
  	 * potential misconfiguration.
 
 *** DIFF OUTPUT TRUNCATED AT 1000 LINES ***
 _______________________________________________
 svn-src-all@freebsd.org mailing list
 http://lists.freebsd.org/mailman/listinfo/svn-src-all
 To unsubscribe, send any mail to "svn-src-all-unsubscribe@freebsd.org"
 
State-Changed-From-To: open->patched 
State-Changed-By: glebius 
State-Changed-When: Fri Dec 16 12:53:17 UTC 2011 
State-Changed-Why:  
In FreeBSD 10 the CARP is no longer an interface, thus this PR isn't applicable. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=120130 
State-Changed-From-To: patched->closed 
State-Changed-By: eadler 
State-Changed-When: Fri Mar 15 23:34:43 UTC 2013 
State-Changed-Why:  
MFCed/fixed by now or it will never be MFCed 

http://www.freebsd.org/cgi/query-pr.cgi?pr=120130 
State-Changed-From-To: closed->patched 
State-Changed-By: glebius 
State-Changed-When: Sat Mar 16 07:31:21 UTC 2013 
State-Changed-Why:  
The PR can be closed only after 10.0-RELEASE is out. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=120130 
State-Changed-From-To: patched->closed 
State-Changed-By: glebius 
State-Changed-When: Wed Jan 22 09:13:56 UTC 2014 
State-Changed-Why:  
Fixed in 10.0-RELEASE. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=120130 
>Unformatted:
