No OneTemporary
Actions

Size

78 KB

Referenced Files

None

Subscribers

None

View Options

	diff --git a/sys/netinet/in_pcb.c b/sys/netinet/in_pcb.c
	index 515bd415e150..9bde3c821242 100644
	--- a/sys/netinet/in_pcb.c
	+++ b/sys/netinet/in_pcb.c
	@@ -1,2100 +1,2128 @@
	/*-
	* Copyright (c) 1982, 1986, 1991, 1993, 1995
	* The Regents of the University of California.
	* Copyright (c) 2007-2009 Robert N. M. Watson
	+ * Copyright (c) 2010-2011 Juniper Networks, Inc.
	* All rights reserved.
	*
	+ * Portions of this software were developed by Robert N. M. Watson under
	+ * contract to Juniper Networks, Inc.
	+ *
	* 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.
	*
	* @(#)in_pcb.c 8.4 (Berkeley) 5/24/95
	*/

	#include <sys/cdefs.h>
	__FBSDID("$FreeBSD$");

	#include "opt_ddb.h"
	#include "opt_ipsec.h"
	#include "opt_inet.h"
	#include "opt_inet6.h"

	#include <sys/param.h>
	#include <sys/systm.h>
	#include <sys/malloc.h>
	#include <sys/mbuf.h>
	#include <sys/callout.h>
	#include <sys/domain.h>
	#include <sys/protosw.h>
	#include <sys/socket.h>
	#include <sys/socketvar.h>
	#include <sys/priv.h>
	#include <sys/proc.h>
	+#include <sys/refcount.h>
	#include <sys/jail.h>
	#include <sys/kernel.h>
	#include <sys/sysctl.h>

	#ifdef DDB
	#include <ddb/ddb.h>
	#endif

	#include <vm/uma.h>

	#include <net/if.h>
	#include <net/if_types.h>
	#include <net/route.h>
	#include <net/vnet.h>

	#if defined(INET) \|\| defined(INET6)
	#include <netinet/in.h>
	#include <netinet/in_pcb.h>
	#include <netinet/ip_var.h>
	#include <netinet/tcp_var.h>
	#include <netinet/udp.h>
	#include <netinet/udp_var.h>
	#endif
	#ifdef INET
	#include <netinet/in_var.h>
	#endif
	#ifdef INET6
	#include <netinet/ip6.h>
	#include <netinet6/in6_pcb.h>
	#include <netinet6/in6_var.h>
	#include <netinet6/ip6_var.h>
	#endif /* INET6 */


	#ifdef IPSEC
	#include <netipsec/ipsec.h>
	#include <netipsec/key.h>
	#endif /* IPSEC */

	#include <security/mac/mac_framework.h>

	static struct callout ipport_tick_callout;

	/*
	* These configure the range of local port addresses assigned to
	* "unspecified" outgoing connections/packets/whatever.
	*/
	VNET_DEFINE(int, ipport_lowfirstauto) = IPPORT_RESERVED - 1; /* 1023 */
	VNET_DEFINE(int, ipport_lowlastauto) = IPPORT_RESERVEDSTART; /* 600 */
	VNET_DEFINE(int, ipport_firstauto) = IPPORT_EPHEMERALFIRST; /* 10000 */
	VNET_DEFINE(int, ipport_lastauto) = IPPORT_EPHEMERALLAST; /* 65535 */
	VNET_DEFINE(int, ipport_hifirstauto) = IPPORT_HIFIRSTAUTO; /* 49152 */
	VNET_DEFINE(int, ipport_hilastauto) = IPPORT_HILASTAUTO; /* 65535 */

	/*
	* Reserved ports accessible only to root. There are significant
	* security considerations that must be accounted for when changing these,
	* but the security benefits can be great. Please be careful.
	*/
	VNET_DEFINE(int, ipport_reservedhigh) = IPPORT_RESERVED - 1; /* 1023 */
	VNET_DEFINE(int, ipport_reservedlow);

	/* Variables dealing with random ephemeral port allocation. */
	VNET_DEFINE(int, ipport_randomized) = 1; /* user controlled via sysctl */
	VNET_DEFINE(int, ipport_randomcps) = 10; /* user controlled via sysctl */
	VNET_DEFINE(int, ipport_randomtime) = 45; /* user controlled via sysctl */
	VNET_DEFINE(int, ipport_stoprandom); /* toggled by ipport_tick */
	VNET_DEFINE(int, ipport_tcpallocs);
	static VNET_DEFINE(int, ipport_tcplastcount);

	#define V_ipport_tcplastcount VNET(ipport_tcplastcount)

	static void in_pcbremlists(struct inpcb *inp);

	#ifdef INET
	#define RANGECHK(var, min, max) \
	if ((var) < (min)) { (var) = (min); } \
	else if ((var) > (max)) { (var) = (max); }

	static int
	sysctl_net_ipport_check(SYSCTL_HANDLER_ARGS)
	{
	int error;

	#ifdef VIMAGE
	error = vnet_sysctl_handle_int(oidp, arg1, arg2, req);
	#else
	error = sysctl_handle_int(oidp, arg1, arg2, req);
	#endif
	if (error == 0) {
	RANGECHK(V_ipport_lowfirstauto, 1, IPPORT_RESERVED - 1);
	RANGECHK(V_ipport_lowlastauto, 1, IPPORT_RESERVED - 1);
	RANGECHK(V_ipport_firstauto, IPPORT_RESERVED, IPPORT_MAX);
	RANGECHK(V_ipport_lastauto, IPPORT_RESERVED, IPPORT_MAX);
	RANGECHK(V_ipport_hifirstauto, IPPORT_RESERVED, IPPORT_MAX);
	RANGECHK(V_ipport_hilastauto, IPPORT_RESERVED, IPPORT_MAX);
	}
	return (error);
	}

	#undef RANGECHK

	SYSCTL_NODE(_net_inet_ip, IPPROTO_IP, portrange, CTLFLAG_RW, 0, "IP Ports");

	SYSCTL_VNET_PROC(_net_inet_ip_portrange, OID_AUTO, lowfirst,
	CTLTYPE_INT\|CTLFLAG_RW, &VNET_NAME(ipport_lowfirstauto), 0,
	&sysctl_net_ipport_check, "I", "");
	SYSCTL_VNET_PROC(_net_inet_ip_portrange, OID_AUTO, lowlast,
	CTLTYPE_INT\|CTLFLAG_RW, &VNET_NAME(ipport_lowlastauto), 0,
	&sysctl_net_ipport_check, "I", "");
	SYSCTL_VNET_PROC(_net_inet_ip_portrange, OID_AUTO, first,
	CTLTYPE_INT\|CTLFLAG_RW, &VNET_NAME(ipport_firstauto), 0,
	&sysctl_net_ipport_check, "I", "");
	SYSCTL_VNET_PROC(_net_inet_ip_portrange, OID_AUTO, last,
	CTLTYPE_INT\|CTLFLAG_RW, &VNET_NAME(ipport_lastauto), 0,
	&sysctl_net_ipport_check, "I", "");
	SYSCTL_VNET_PROC(_net_inet_ip_portrange, OID_AUTO, hifirst,
	CTLTYPE_INT\|CTLFLAG_RW, &VNET_NAME(ipport_hifirstauto), 0,
	&sysctl_net_ipport_check, "I", "");
	SYSCTL_VNET_PROC(_net_inet_ip_portrange, OID_AUTO, hilast,
	CTLTYPE_INT\|CTLFLAG_RW, &VNET_NAME(ipport_hilastauto), 0,
	&sysctl_net_ipport_check, "I", "");
	SYSCTL_VNET_INT(_net_inet_ip_portrange, OID_AUTO, reservedhigh,
	CTLFLAG_RW\|CTLFLAG_SECURE, &VNET_NAME(ipport_reservedhigh), 0, "");
	SYSCTL_VNET_INT(_net_inet_ip_portrange, OID_AUTO, reservedlow,
	CTLFLAG_RW\|CTLFLAG_SECURE, &VNET_NAME(ipport_reservedlow), 0, "");
	SYSCTL_VNET_INT(_net_inet_ip_portrange, OID_AUTO, randomized, CTLFLAG_RW,
	&VNET_NAME(ipport_randomized), 0, "Enable random port allocation");
	SYSCTL_VNET_INT(_net_inet_ip_portrange, OID_AUTO, randomcps, CTLFLAG_RW,
	&VNET_NAME(ipport_randomcps), 0, "Maximum number of random port "
	"allocations before switching to a sequental one");
	SYSCTL_VNET_INT(_net_inet_ip_portrange, OID_AUTO, randomtime, CTLFLAG_RW,
	&VNET_NAME(ipport_randomtime), 0,
	"Minimum time to keep sequental port "
	"allocation before switching to a random one");
	#endif

	/*
	* in_pcb.c: manage the Protocol Control Blocks.
	*
	* NOTE: It is assumed that most of these functions will be called with
	* the pcbinfo lock held, and often, the inpcb lock held, as these utility
	* functions often modify hash chains or addresses in pcbs.
	*/

	/*
	* Initialize an inpcbinfo -- we should be able to reduce the number of
	* arguments in time.
	*/
	void
	in_pcbinfo_init(struct inpcbinfo pcbinfo, const char name,
	struct inpcbhead *listhead, int hash_nelements, int porthash_nelements,
	char *inpcbzone_name, uma_init inpcbzone_init, uma_fini inpcbzone_fini,
	uint32_t inpcbzone_flags)
	{

	INP_INFO_LOCK_INIT(pcbinfo, name);
	#ifdef VIMAGE
	pcbinfo->ipi_vnet = curvnet;
	#endif
	pcbinfo->ipi_listhead = listhead;
	LIST_INIT(pcbinfo->ipi_listhead);
	pcbinfo->ipi_hashbase = hashinit(hash_nelements, M_PCB,
	&pcbinfo->ipi_hashmask);
	pcbinfo->ipi_porthashbase = hashinit(porthash_nelements, M_PCB,
	&pcbinfo->ipi_porthashmask);
	pcbinfo->ipi_zone = uma_zcreate(inpcbzone_name, sizeof(struct inpcb),
	NULL, NULL, inpcbzone_init, inpcbzone_fini, UMA_ALIGN_PTR,
	inpcbzone_flags);
	uma_zone_set_max(pcbinfo->ipi_zone, maxsockets);
	}

	/*
	* Destroy an inpcbinfo.
	*/
	void
	in_pcbinfo_destroy(struct inpcbinfo *pcbinfo)
	{

	hashdestroy(pcbinfo->ipi_hashbase, M_PCB, pcbinfo->ipi_hashmask);
	hashdestroy(pcbinfo->ipi_porthashbase, M_PCB,
	pcbinfo->ipi_porthashmask);
	uma_zdestroy(pcbinfo->ipi_zone);
	INP_INFO_LOCK_DESTROY(pcbinfo);
	}

	/*
	* Allocate a PCB and associate it with the socket.
	* On success return with the PCB locked.
	*/
	int
	in_pcballoc(struct socket so, struct inpcbinfo pcbinfo)
	{
	struct inpcb *inp;
	int error;

	INP_INFO_WLOCK_ASSERT(pcbinfo);
	error = 0;
	inp = uma_zalloc(pcbinfo->ipi_zone, M_NOWAIT);
	if (inp == NULL)
	return (ENOBUFS);
	bzero(inp, inp_zero_size);
	inp->inp_pcbinfo = pcbinfo;
	inp->inp_socket = so;
	inp->inp_cred = crhold(so->so_cred);
	inp->inp_inc.inc_fibnum = so->so_fibnum;
	#ifdef MAC
	error = mac_inpcb_init(inp, M_NOWAIT);
	if (error != 0)
	goto out;
	mac_inpcb_create(so, inp);
	#endif
	#ifdef IPSEC
	error = ipsec_init_policy(so, &inp->inp_sp);
	if (error != 0) {
	#ifdef MAC
	mac_inpcb_destroy(inp);
	#endif
	goto out;
	}
	#endif /IPSEC/
	#ifdef INET6
	if (INP_SOCKAF(so) == AF_INET6) {
	inp->inp_vflag \|= INP_IPV6PROTO;
	if (V_ip6_v6only)
	inp->inp_flags \|= IN6P_IPV6_V6ONLY;
	}
	#endif
	LIST_INSERT_HEAD(pcbinfo->ipi_listhead, inp, inp_list);
	pcbinfo->ipi_count++;
	so->so_pcb = (caddr_t)inp;
	#ifdef INET6
	if (V_ip6_auto_flowlabel)
	inp->inp_flags \|= IN6P_AUTOFLOWLABEL;
	#endif
	INP_WLOCK(inp);
	inp->inp_gencnt = ++pcbinfo->ipi_gencnt;
	- inp->inp_refcount = 1; /* Reference from the inpcbinfo */
	+ refcount_init(&inp->inp_refcount, 1); /* Reference from inpcbinfo */
	#if defined(IPSEC) \|\| defined(MAC)
	out:
	if (error != 0) {
	crfree(inp->inp_cred);
	uma_zfree(pcbinfo->ipi_zone, inp);
	}
	#endif
	return (error);
	}

	#ifdef INET
	int
	in_pcbbind(struct inpcb inp, struct sockaddr nam, struct ucred *cred)
	{
	int anonport, error;

	INP_INFO_WLOCK_ASSERT(inp->inp_pcbinfo);
	INP_WLOCK_ASSERT(inp);

	if (inp->inp_lport != 0 \|\| inp->inp_laddr.s_addr != INADDR_ANY)
	return (EINVAL);
	anonport = inp->inp_lport == 0 && (nam == NULL \|\|
	((struct sockaddr_in *)nam)->sin_port == 0);
	error = in_pcbbind_setup(inp, nam, &inp->inp_laddr.s_addr,
	&inp->inp_lport, cred);
	if (error)
	return (error);
	if (in_pcbinshash(inp) != 0) {
	inp->inp_laddr.s_addr = INADDR_ANY;
	inp->inp_lport = 0;
	return (EAGAIN);
	}
	if (anonport)
	inp->inp_flags \|= INP_ANONPORT;
	return (0);
	}
	#endif

	#if defined(INET) \|\| defined(INET6)
	int
	in_pcb_lport(struct inpcb inp, struct in_addr laddrp, u_short *lportp,
	struct ucred *cred, int lookupflags)
	{
	struct inpcbinfo *pcbinfo;
	struct inpcb *tmpinp;
	unsigned short *lastport;
	int count, dorandom, error;
	u_short aux, first, last, lport;
	#ifdef INET
	struct in_addr laddr;
	#endif

	pcbinfo = inp->inp_pcbinfo;

	/*
	* Because no actual state changes occur here, a global write lock on
	* the pcbinfo isn't required.
	*/
	INP_INFO_LOCK_ASSERT(pcbinfo);
	INP_LOCK_ASSERT(inp);

	if (inp->inp_flags & INP_HIGHPORT) {
	first = V_ipport_hifirstauto; /* sysctl */
	last = V_ipport_hilastauto;
	lastport = &pcbinfo->ipi_lasthi;
	} else if (inp->inp_flags & INP_LOWPORT) {
	error = priv_check_cred(cred, PRIV_NETINET_RESERVEDPORT, 0);
	if (error)
	return (error);
	first = V_ipport_lowfirstauto; /* 1023 */
	last = V_ipport_lowlastauto; /* 600 */
	lastport = &pcbinfo->ipi_lastlow;
	} else {
	first = V_ipport_firstauto; /* sysctl */
	last = V_ipport_lastauto;
	lastport = &pcbinfo->ipi_lastport;
	}
	/*
	* For UDP, use random port allocation as long as the user
	* allows it. For TCP (and as of yet unknown) connections,
	* use random port allocation only if the user allows it AND
	* ipport_tick() allows it.
	*/
	if (V_ipport_randomized &&
	(!V_ipport_stoprandom \|\| pcbinfo == &V_udbinfo))
	dorandom = 1;
	else
	dorandom = 0;
	/*
	* It makes no sense to do random port allocation if
	* we have the only port available.
	*/
	if (first == last)
	dorandom = 0;
	/* Make sure to not include UDP packets in the count. */
	if (pcbinfo != &V_udbinfo)
	V_ipport_tcpallocs++;
	/*
	* Instead of having two loops further down counting up or down
	* make sure that first is always <= last and go with only one
	* code path implementing all logic.
	*/
	if (first > last) {
	aux = first;
	first = last;
	last = aux;
	}

	#ifdef INET
	/* Make the compiler happy. */
	laddr.s_addr = 0;
	if ((inp->inp_vflag & (INP_IPV4\|INP_IPV6)) == INP_IPV4) {
	KASSERT(laddrp != NULL, ("%s: laddrp NULL for v4 inp %p",
	__func__, inp));
	laddr = *laddrp;
	}
	#endif
	tmpinp = NULL; /* Make compiler happy. */
	lport = *lportp;

	if (dorandom)
	*lastport = first + (arc4random() % (last - first));

	count = last - first;

	do {
	if (count-- < 0) /* completely used? */
	return (EADDRNOTAVAIL);
	++*lastport;
	if (lastport < first \|\| lastport > last)
	*lastport = first;
	lport = htons(*lastport);

	#ifdef INET6
	if ((inp->inp_vflag & INP_IPV6) != 0)
	tmpinp = in6_pcblookup_local(pcbinfo,
	&inp->in6p_laddr, lport, lookupflags, cred);
	#endif
	#if defined(INET) && defined(INET6)
	else
	#endif
	#ifdef INET
	tmpinp = in_pcblookup_local(pcbinfo, laddr,
	lport, lookupflags, cred);
	#endif
	} while (tmpinp != NULL);

	#ifdef INET
	if ((inp->inp_vflag & (INP_IPV4\|INP_IPV6)) == INP_IPV4)
	laddrp->s_addr = laddr.s_addr;
	#endif
	*lportp = lport;

	return (0);
	}
	#endif /* INET \|\| INET6 */

	#ifdef INET
	/*
	* Set up a bind operation on a PCB, performing port allocation
	* as required, but do not actually modify the PCB. Callers can
	* either complete the bind by setting inp_laddr/inp_lport and
	* calling in_pcbinshash(), or they can just use the resulting
	* port and address to authorise the sending of a once-off packet.
	*
	* On error, the values of laddrp and lportp are not changed.
	*/
	int
	in_pcbbind_setup(struct inpcb inp, struct sockaddr nam, in_addr_t *laddrp,
	u_short lportp, struct ucred cred)
	{
	struct socket *so = inp->inp_socket;
	struct sockaddr_in *sin;
	struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;
	struct in_addr laddr;
	u_short lport = 0;
	int lookupflags = 0, reuseport = (so->so_options & SO_REUSEPORT);
	int error;

	/*
	* Because no actual state changes occur here, a global write lock on
	* the pcbinfo isn't required.
	*/
	INP_INFO_LOCK_ASSERT(pcbinfo);
	INP_LOCK_ASSERT(inp);

	if (TAILQ_EMPTY(&V_in_ifaddrhead)) /* XXX broken! */
	return (EADDRNOTAVAIL);
	laddr.s_addr = *laddrp;
	if (nam != NULL && laddr.s_addr != INADDR_ANY)
	return (EINVAL);
	if ((so->so_options & (SO_REUSEADDR\|SO_REUSEPORT)) == 0)
	lookupflags = INPLOOKUP_WILDCARD;
	if (nam == NULL) {
	if ((error = prison_local_ip4(cred, &laddr)) != 0)
	return (error);
	} else {
	sin = (struct sockaddr_in *)nam;
	if (nam->sa_len != sizeof (*sin))
	return (EINVAL);
	#ifdef notdef
	/*
	* We should check the family, but old programs
	* incorrectly fail to initialize it.
	*/
	if (sin->sin_family != AF_INET)
	return (EAFNOSUPPORT);
	#endif
	error = prison_local_ip4(cred, &sin->sin_addr);
	if (error)
	return (error);
	if (sin->sin_port != *lportp) {
	/* Don't allow the port to change. */
	if (*lportp != 0)
	return (EINVAL);
	lport = sin->sin_port;
	}
	/* NB: lport is left as 0 if the port isn't being changed. */
	if (IN_MULTICAST(ntohl(sin->sin_addr.s_addr))) {
	/*
	* Treat SO_REUSEADDR as SO_REUSEPORT for multicast;
	* allow complete duplication of binding if
	* SO_REUSEPORT is set, or if SO_REUSEADDR is set
	* and a multicast address is bound on both
	* new and duplicated sockets.
	*/
	if (so->so_options & SO_REUSEADDR)
	reuseport = SO_REUSEADDR\|SO_REUSEPORT;
	} else if (sin->sin_addr.s_addr != INADDR_ANY) {
	sin->sin_port = 0; /* yech... */
	bzero(&sin->sin_zero, sizeof(sin->sin_zero));
	/*
	* Is the address a local IP address?
	* If INP_BINDANY is set, then the socket may be bound
	* to any endpoint address, local or not.
	*/
	if ((inp->inp_flags & INP_BINDANY) == 0 &&
	ifa_ifwithaddr_check((struct sockaddr *)sin) == 0)
	return (EADDRNOTAVAIL);
	}
	laddr = sin->sin_addr;
	if (lport) {
	struct inpcb *t;
	struct tcptw *tw;

	/* GROSS */
	if (ntohs(lport) <= V_ipport_reservedhigh &&
	ntohs(lport) >= V_ipport_reservedlow &&
	priv_check_cred(cred, PRIV_NETINET_RESERVEDPORT,
	0))
	return (EACCES);
	if (!IN_MULTICAST(ntohl(sin->sin_addr.s_addr)) &&
	priv_check_cred(inp->inp_cred,
	PRIV_NETINET_REUSEPORT, 0) != 0) {
	t = in_pcblookup_local(pcbinfo, sin->sin_addr,
	lport, INPLOOKUP_WILDCARD, cred);
	/*
	* XXX
	* This entire block sorely needs a rewrite.
	*/
	if (t &&
	((t->inp_flags & INP_TIMEWAIT) == 0) &&
	(so->so_type != SOCK_STREAM \|\|
	ntohl(t->inp_faddr.s_addr) == INADDR_ANY) &&
	(ntohl(sin->sin_addr.s_addr) != INADDR_ANY \|\|
	ntohl(t->inp_laddr.s_addr) != INADDR_ANY \|\|
	(t->inp_socket->so_options &
	SO_REUSEPORT) == 0) &&
	(inp->inp_cred->cr_uid !=
	t->inp_cred->cr_uid))
	return (EADDRINUSE);
	}
	t = in_pcblookup_local(pcbinfo, sin->sin_addr,
	lport, lookupflags, cred);
	if (t && (t->inp_flags & INP_TIMEWAIT)) {
	/*
	* XXXRW: If an incpb has had its timewait
	* state recycled, we treat the address as
	* being in use (for now). This is better
	* than a panic, but not desirable.
	*/
	tw = intotw(inp);
	if (tw == NULL \|\|
	(reuseport & tw->tw_so_options) == 0)
	return (EADDRINUSE);
	} else if (t &&
	(reuseport & t->inp_socket->so_options) == 0) {
	#ifdef INET6
	if (ntohl(sin->sin_addr.s_addr) !=
	INADDR_ANY \|\|
	ntohl(t->inp_laddr.s_addr) !=
	INADDR_ANY \|\|
	INP_SOCKAF(so) ==
	INP_SOCKAF(t->inp_socket))
	#endif
	return (EADDRINUSE);
	}
	}
	}
	if (*lportp != 0)
	lport = *lportp;
	if (lport == 0) {
	error = in_pcb_lport(inp, &laddr, &lport, cred, lookupflags);
	if (error != 0)
	return (error);

	}
	*laddrp = laddr.s_addr;
	*lportp = lport;
	return (0);
	}

	/*
	* Connect from a socket to a specified address.
	* Both address and port must be specified in argument sin.
	* If don't have a local address for this socket yet,
	* then pick one.
	*/
	int
	in_pcbconnect(struct inpcb inp, struct sockaddr nam, struct ucred *cred)
	{
	u_short lport, fport;
	in_addr_t laddr, faddr;
	int anonport, error;

	INP_INFO_WLOCK_ASSERT(inp->inp_pcbinfo);
	INP_WLOCK_ASSERT(inp);

	lport = inp->inp_lport;
	laddr = inp->inp_laddr.s_addr;
	anonport = (lport == 0);
	error = in_pcbconnect_setup(inp, nam, &laddr, &lport, &faddr, &fport,
	NULL, cred);
	if (error)
	return (error);

	/* Do the initial binding of the local address if required. */
	if (inp->inp_laddr.s_addr == INADDR_ANY && inp->inp_lport == 0) {
	inp->inp_lport = lport;
	inp->inp_laddr.s_addr = laddr;
	if (in_pcbinshash(inp) != 0) {
	inp->inp_laddr.s_addr = INADDR_ANY;
	inp->inp_lport = 0;
	return (EAGAIN);
	}
	}

	/* Commit the remaining changes. */
	inp->inp_lport = lport;
	inp->inp_laddr.s_addr = laddr;
	inp->inp_faddr.s_addr = faddr;
	inp->inp_fport = fport;
	in_pcbrehash(inp);

	if (anonport)
	inp->inp_flags \|= INP_ANONPORT;
	return (0);
	}

	/*
	* Do proper source address selection on an unbound socket in case
	* of connect. Take jails into account as well.
	*/
	static int
	in_pcbladdr(struct inpcb inp, struct in_addr faddr, struct in_addr *laddr,
	struct ucred *cred)
	{
	struct ifaddr *ifa;
	struct sockaddr *sa;
	struct sockaddr_in *sin;
	struct route sro;
	int error;

	KASSERT(laddr != NULL, ("%s: laddr NULL", __func__));

	/*
	* Bypass source address selection and use the primary jail IP
	* if requested.
	*/
	if (cred != NULL && !prison_saddrsel_ip4(cred, laddr))
	return (0);

	error = 0;
	bzero(&sro, sizeof(sro));

	sin = (struct sockaddr_in *)&sro.ro_dst;
	sin->sin_family = AF_INET;
	sin->sin_len = sizeof(struct sockaddr_in);
	sin->sin_addr.s_addr = faddr->s_addr;

	/*
	* If route is known our src addr is taken from the i/f,
	* else punt.
	*
	* Find out route to destination.
	*/
	if ((inp->inp_socket->so_options & SO_DONTROUTE) == 0)
	in_rtalloc_ign(&sro, 0, inp->inp_inc.inc_fibnum);

	/*
	* If we found a route, use the address corresponding to
	* the outgoing interface.
	*
	* Otherwise assume faddr is reachable on a directly connected
	* network and try to find a corresponding interface to take
	* the source address from.
	*/
	if (sro.ro_rt == NULL \|\| sro.ro_rt->rt_ifp == NULL) {
	struct in_ifaddr *ia;
	struct ifnet *ifp;

	ia = ifatoia(ifa_ifwithdstaddr((struct sockaddr *)sin));
	if (ia == NULL)
	ia = ifatoia(ifa_ifwithnet((struct sockaddr *)sin, 0));
	if (ia == NULL) {
	error = ENETUNREACH;
	goto done;
	}

	if (cred == NULL \|\| !prison_flag(cred, PR_IP4)) {
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	ifa_free(&ia->ia_ifa);
	goto done;
	}

	ifp = ia->ia_ifp;
	ifa_free(&ia->ia_ifa);
	ia = NULL;
	IF_ADDR_LOCK(ifp);
	TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {

	sa = ifa->ifa_addr;
	if (sa->sa_family != AF_INET)
	continue;
	sin = (struct sockaddr_in *)sa;
	if (prison_check_ip4(cred, &sin->sin_addr) == 0) {
	ia = (struct in_ifaddr *)ifa;
	break;
	}
	}
	if (ia != NULL) {
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	IF_ADDR_UNLOCK(ifp);
	goto done;
	}
	IF_ADDR_UNLOCK(ifp);

	/* 3. As a last resort return the 'default' jail address. */
	error = prison_get_ip4(cred, laddr);
	goto done;
	}

	/*
	* If the outgoing interface on the route found is not
	* a loopback interface, use the address from that interface.
	* In case of jails do those three steps:
	* 1. check if the interface address belongs to the jail. If so use it.
	* 2. check if we have any address on the outgoing interface
	* belonging to this jail. If so use it.
	* 3. as a last resort return the 'default' jail address.
	*/
	if ((sro.ro_rt->rt_ifp->if_flags & IFF_LOOPBACK) == 0) {
	struct in_ifaddr *ia;
	struct ifnet *ifp;

	/* If not jailed, use the default returned. */
	if (cred == NULL \|\| !prison_flag(cred, PR_IP4)) {
	ia = (struct in_ifaddr *)sro.ro_rt->rt_ifa;
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	goto done;
	}

	/* Jailed. */
	/* 1. Check if the iface address belongs to the jail. */
	sin = (struct sockaddr_in *)sro.ro_rt->rt_ifa->ifa_addr;
	if (prison_check_ip4(cred, &sin->sin_addr) == 0) {
	ia = (struct in_ifaddr *)sro.ro_rt->rt_ifa;
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	goto done;
	}

	/*
	* 2. Check if we have any address on the outgoing interface
	* belonging to this jail.
	*/
	ia = NULL;
	ifp = sro.ro_rt->rt_ifp;
	IF_ADDR_LOCK(ifp);
	TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
	sa = ifa->ifa_addr;
	if (sa->sa_family != AF_INET)
	continue;
	sin = (struct sockaddr_in *)sa;
	if (prison_check_ip4(cred, &sin->sin_addr) == 0) {
	ia = (struct in_ifaddr *)ifa;
	break;
	}
	}
	if (ia != NULL) {
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	IF_ADDR_UNLOCK(ifp);
	goto done;
	}
	IF_ADDR_UNLOCK(ifp);

	/* 3. As a last resort return the 'default' jail address. */
	error = prison_get_ip4(cred, laddr);
	goto done;
	}

	/*
	* The outgoing interface is marked with 'loopback net', so a route
	* to ourselves is here.
	* Try to find the interface of the destination address and then
	* take the address from there. That interface is not necessarily
	* a loopback interface.
	* In case of jails, check that it is an address of the jail
	* and if we cannot find, fall back to the 'default' jail address.
	*/
	if ((sro.ro_rt->rt_ifp->if_flags & IFF_LOOPBACK) != 0) {
	struct sockaddr_in sain;
	struct in_ifaddr *ia;

	bzero(&sain, sizeof(struct sockaddr_in));
	sain.sin_family = AF_INET;
	sain.sin_len = sizeof(struct sockaddr_in);
	sain.sin_addr.s_addr = faddr->s_addr;

	ia = ifatoia(ifa_ifwithdstaddr(sintosa(&sain)));
	if (ia == NULL)
	ia = ifatoia(ifa_ifwithnet(sintosa(&sain), 0));
	if (ia == NULL)
	ia = ifatoia(ifa_ifwithaddr(sintosa(&sain)));

	if (cred == NULL \|\| !prison_flag(cred, PR_IP4)) {
	if (ia == NULL) {
	error = ENETUNREACH;
	goto done;
	}
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	ifa_free(&ia->ia_ifa);
	goto done;
	}

	/* Jailed. */
	if (ia != NULL) {
	struct ifnet *ifp;

	ifp = ia->ia_ifp;
	ifa_free(&ia->ia_ifa);
	ia = NULL;
	IF_ADDR_LOCK(ifp);
	TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {

	sa = ifa->ifa_addr;
	if (sa->sa_family != AF_INET)
	continue;
	sin = (struct sockaddr_in *)sa;
	if (prison_check_ip4(cred,
	&sin->sin_addr) == 0) {
	ia = (struct in_ifaddr *)ifa;
	break;
	}
	}
	if (ia != NULL) {
	laddr->s_addr = ia->ia_addr.sin_addr.s_addr;
	IF_ADDR_UNLOCK(ifp);
	goto done;
	}
	IF_ADDR_UNLOCK(ifp);
	}

	/* 3. As a last resort return the 'default' jail address. */
	error = prison_get_ip4(cred, laddr);
	goto done;
	}

	done:
	if (sro.ro_rt != NULL)
	RTFREE(sro.ro_rt);
	return (error);
	}

	/*
	* Set up for a connect from a socket to the specified address.
	* On entry, laddrp and lportp should contain the current local
	* address and port for the PCB; these are updated to the values
	* that should be placed in inp_laddr and inp_lport to complete
	* the connect.
	*
	* On success, faddrp and fportp will be set to the remote address
	* and port. These are not updated in the error case.
	*
	* If the operation fails because the connection already exists,
	* *oinpp will be set to the PCB of that connection so that the
	* caller can decide to override it. In all other cases, *oinpp
	* is set to NULL.
	*/
	int
	in_pcbconnect_setup(struct inpcb inp, struct sockaddr nam,
	in_addr_t laddrp, u_short lportp, in_addr_t faddrp, u_short fportp,
	struct inpcb *oinpp, struct ucred cred)
	{
	struct sockaddr_in sin = (struct sockaddr_in )nam;
	struct in_ifaddr *ia;
	struct inpcb *oinp;
	struct in_addr laddr, faddr;
	u_short lport, fport;
	int error;

	/*
	* Because a global state change doesn't actually occur here, a read
	* lock is sufficient.
	*/
	INP_INFO_LOCK_ASSERT(inp->inp_pcbinfo);
	INP_LOCK_ASSERT(inp);

	if (oinpp != NULL)
	*oinpp = NULL;
	if (nam->sa_len != sizeof (*sin))
	return (EINVAL);
	if (sin->sin_family != AF_INET)
	return (EAFNOSUPPORT);
	if (sin->sin_port == 0)
	return (EADDRNOTAVAIL);
	laddr.s_addr = *laddrp;
	lport = *lportp;
	faddr = sin->sin_addr;
	fport = sin->sin_port;

	if (!TAILQ_EMPTY(&V_in_ifaddrhead)) {
	/*
	* If the destination address is INADDR_ANY,
	* use the primary local address.
	* If the supplied address is INADDR_BROADCAST,
	* and the primary interface supports broadcast,
	* choose the broadcast address for that interface.
	*/
	if (faddr.s_addr == INADDR_ANY) {
	IN_IFADDR_RLOCK();
	faddr =
	IA_SIN(TAILQ_FIRST(&V_in_ifaddrhead))->sin_addr;
	IN_IFADDR_RUNLOCK();
	if (cred != NULL &&
	(error = prison_get_ip4(cred, &faddr)) != 0)
	return (error);
	} else if (faddr.s_addr == (u_long)INADDR_BROADCAST) {
	IN_IFADDR_RLOCK();
	if (TAILQ_FIRST(&V_in_ifaddrhead)->ia_ifp->if_flags &
	IFF_BROADCAST)
	faddr = satosin(&TAILQ_FIRST(
	&V_in_ifaddrhead)->ia_broadaddr)->sin_addr;
	IN_IFADDR_RUNLOCK();
	}
	}
	if (laddr.s_addr == INADDR_ANY) {
	error = in_pcbladdr(inp, &faddr, &laddr, cred);
	/*
	* If the destination address is multicast and an outgoing
	* interface has been set as a multicast option, prefer the
	* address of that interface as our source address.
	*/
	if (IN_MULTICAST(ntohl(faddr.s_addr)) &&
	inp->inp_moptions != NULL) {
	struct ip_moptions *imo;
	struct ifnet *ifp;

	imo = inp->inp_moptions;
	if (imo->imo_multicast_ifp != NULL) {
	ifp = imo->imo_multicast_ifp;
	IN_IFADDR_RLOCK();
	TAILQ_FOREACH(ia, &V_in_ifaddrhead, ia_link) {
	if ((ia->ia_ifp == ifp) &&
	(cred == NULL \|\|
	prison_check_ip4(cred,
	&ia->ia_addr.sin_addr) == 0))
	break;
	}
	if (ia == NULL)
	error = EADDRNOTAVAIL;
	else {
	laddr = ia->ia_addr.sin_addr;
	error = 0;
	}
	IN_IFADDR_RUNLOCK();
	}
	}
	if (error)
	return (error);
	}
	oinp = in_pcblookup_hash(inp->inp_pcbinfo, faddr, fport, laddr, lport,
	0, NULL);
	if (oinp != NULL) {
	if (oinpp != NULL)
	*oinpp = oinp;
	return (EADDRINUSE);
	}
	if (lport == 0) {
	error = in_pcbbind_setup(inp, NULL, &laddr.s_addr, &lport,
	cred);
	if (error)
	return (error);
	}
	*laddrp = laddr.s_addr;
	*lportp = lport;
	*faddrp = faddr.s_addr;
	*fportp = fport;
	return (0);
	}

	void
	in_pcbdisconnect(struct inpcb *inp)
	{

	INP_INFO_WLOCK_ASSERT(inp->inp_pcbinfo);
	INP_WLOCK_ASSERT(inp);

	inp->inp_faddr.s_addr = INADDR_ANY;
	inp->inp_fport = 0;
	in_pcbrehash(inp);
	}
	#endif

	/*
	* in_pcbdetach() is responsibe for disassociating a socket from an inpcb.
	* For most protocols, this will be invoked immediately prior to calling
	* in_pcbfree(). However, with TCP the inpcb may significantly outlive the
	* socket, in which case in_pcbfree() is deferred.
	*/
	void
	in_pcbdetach(struct inpcb *inp)
	{

	KASSERT(inp->inp_socket != NULL, ("%s: inp_socket == NULL", __func__));

	inp->inp_socket->so_pcb = NULL;
	inp->inp_socket = NULL;
	}

	-/*
	- * in_pcbfree_internal() frees an inpcb that has been detached from its
	- * socket, and whose reference count has reached 0. It will also remove the
	- * inpcb from any global lists it might remain on.
	- */
	-static void
	-in_pcbfree_internal(struct inpcb *inp)
	-{
	- struct inpcbinfo *ipi = inp->inp_pcbinfo;
	-
	- KASSERT(inp->inp_socket == NULL, ("%s: inp_socket != NULL", __func__));
	- KASSERT(inp->inp_refcount == 0, ("%s: refcount !0", __func__));
	-
	- INP_INFO_WLOCK_ASSERT(ipi);
	- INP_WLOCK_ASSERT(inp);
	-
	-#ifdef IPSEC
	- if (inp->inp_sp != NULL)
	- ipsec_delete_pcbpolicy(inp);
	-#endif /* IPSEC */
	- inp->inp_gencnt = ++ipi->ipi_gencnt;
	- in_pcbremlists(inp);
	-#ifdef INET6
	- if (inp->inp_vflag & INP_IPV6PROTO) {
	- ip6_freepcbopts(inp->in6p_outputopts);
	- if (inp->in6p_moptions != NULL)
	- ip6_freemoptions(inp->in6p_moptions);
	- }
	-#endif
	- if (inp->inp_options)
	- (void)m_free(inp->inp_options);
	-#ifdef INET
	- if (inp->inp_moptions != NULL)
	- inp_freemoptions(inp->inp_moptions);
	-#endif
	- inp->inp_vflag = 0;
	- crfree(inp->inp_cred);
	-
	-#ifdef MAC
	- mac_inpcb_destroy(inp);
	-#endif
	- INP_WUNLOCK(inp);
	- uma_zfree(ipi->ipi_zone, inp);
	-}
	-
	/*
	* in_pcbref() bumps the reference count on an inpcb in order to maintain
	* stability of an inpcb pointer despite the inpcb lock being released. This
	* is used in TCP when the inpcbinfo lock needs to be acquired or upgraded,
	* but where the inpcb lock is already held.
	*
	+ * in_pcbref() should be used only to provide brief memory stability, and
	+ * must always be followed by a call to INP_WLOCK() and in_pcbrele() to
	+ * garbage collect the inpcb if it has been in_pcbfree()'d from another
	+ * context. Until in_pcbrele() has returned that the inpcb is still valid,
	+ * lock and rele are the only safe operations that may be performed on the
	+ * inpcb.
	+ *
	* While the inpcb will not be freed, releasing the inpcb lock means that the
	* connection's state may change, so the caller should be careful to
	* revalidate any cached state on reacquiring the lock. Drop the reference
	* using in_pcbrele().
	*/
	void
	in_pcbref(struct inpcb *inp)
	{

	INP_WLOCK_ASSERT(inp);

	KASSERT(inp->inp_refcount > 0, ("%s: refcount 0", __func__));

	- inp->inp_refcount++;
	+ refcount_acquire(&inp->inp_refcount);
	}

	/*
	* Drop a refcount on an inpcb elevated using in_pcbref(); because a call to
	* in_pcbfree() may have been made between in_pcbref() and in_pcbrele(), we
	* return a flag indicating whether or not the inpcb remains valid. If it is
	* valid, we return with the inpcb lock held.
	+ *
	+ * Notice that, unlike in_pcbref(), the inpcb lock must be held to drop a
	+ * reference on an inpcb. Historically more work was done here (actually, in
	+ * in_pcbfree_internal()) but has been moved to in_pcbfree() to avoid the
	+ * need for the pcbinfo lock in in_pcbrele(). Deferring the free is entirely
	+ * about memory stability (and continued use of the write lock).
	*/
	int
	-in_pcbrele(struct inpcb *inp)
	+in_pcbrele_rlocked(struct inpcb *inp)
	{
	-#ifdef INVARIANTS
	- struct inpcbinfo *ipi = inp->inp_pcbinfo;
	-#endif
	+ struct inpcbinfo *pcbinfo;
	+
	+ KASSERT(inp->inp_refcount > 0, ("%s: refcount 0", __func__));
	+
	+ INP_RLOCK_ASSERT(inp);
	+
	+ if (refcount_release(&inp->inp_refcount) == 0)
	+ return (0);
	+
	+ KASSERT(inp->inp_socket == NULL, ("%s: inp_socket != NULL", __func__));
	+
	+ INP_RUNLOCK(inp);
	+ pcbinfo = inp->inp_pcbinfo;
	+ uma_zfree(pcbinfo->ipi_zone, inp);
	+ return (1);
	+}
	+
	+int
	+in_pcbrele_wlocked(struct inpcb *inp)
	+{
	+ struct inpcbinfo *pcbinfo;

	KASSERT(inp->inp_refcount > 0, ("%s: refcount 0", __func__));

	- INP_INFO_WLOCK_ASSERT(ipi);
	INP_WLOCK_ASSERT(inp);

	- inp->inp_refcount--;
	- if (inp->inp_refcount > 0)
	+ if (refcount_release(&inp->inp_refcount) == 0)
	return (0);
	- in_pcbfree_internal(inp);
	+
	+ KASSERT(inp->inp_socket == NULL, ("%s: inp_socket != NULL", __func__));
	+
	+ INP_WUNLOCK(inp);
	+ pcbinfo = inp->inp_pcbinfo;
	+ uma_zfree(pcbinfo->ipi_zone, inp);
	return (1);
	}

	+/*
	+ * Temporary wrapper.
	+ */
	+int
	+in_pcbrele(struct inpcb *inp)
	+{
	+
	+ return (in_pcbrele_wlocked(inp));
	+}
	+
	/*
	* Unconditionally schedule an inpcb to be freed by decrementing its
	* reference count, which should occur only after the inpcb has been detached
	* from its socket. If another thread holds a temporary reference (acquired
	* using in_pcbref()) then the free is deferred until that reference is
	- * released using in_pcbrele(), but the inpcb is still unlocked.
	+ * released using in_pcbrele(), but the inpcb is still unlocked. Almost all
	+ * work, including removal from global lists, is done in this context, where
	+ * the pcbinfo lock is held.
	*/
	void
	in_pcbfree(struct inpcb *inp)
	{
	-#ifdef INVARIANTS
	- struct inpcbinfo *ipi = inp->inp_pcbinfo;
	-#endif
	+ struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;

	- KASSERT(inp->inp_socket == NULL, ("%s: inp_socket != NULL",
	- __func__));
	+ KASSERT(inp->inp_socket == NULL, ("%s: inp_socket != NULL", __func__));

	- INP_INFO_WLOCK_ASSERT(ipi);
	+ INP_INFO_WLOCK_ASSERT(pcbinfo);
	INP_WLOCK_ASSERT(inp);

	- if (!in_pcbrele(inp))
	+ /* XXXRW: Do as much as possible here. */
	+#ifdef IPSEC
	+ if (inp->inp_sp != NULL)
	+ ipsec_delete_pcbpolicy(inp);
	+#endif /* IPSEC */
	+ inp->inp_gencnt = ++pcbinfo->ipi_gencnt;
	+ in_pcbremlists(inp);
	+#ifdef INET6
	+ if (inp->inp_vflag & INP_IPV6PROTO) {
	+ ip6_freepcbopts(inp->in6p_outputopts);
	+ if (inp->in6p_moptions != NULL)
	+ ip6_freemoptions(inp->in6p_moptions);
	+ }
	+#endif
	+ if (inp->inp_options)
	+ (void)m_free(inp->inp_options);
	+#ifdef INET
	+ if (inp->inp_moptions != NULL)
	+ inp_freemoptions(inp->inp_moptions);
	+#endif
	+ inp->inp_vflag = 0;
	+ crfree(inp->inp_cred);
	+#ifdef MAC
	+ mac_inpcb_destroy(inp);
	+#endif
	+ if (!in_pcbrele_wlocked(inp))
	INP_WUNLOCK(inp);
	}

	/*
	* in_pcbdrop() removes an inpcb from hashed lists, releasing its address and
	* port reservation, and preventing it from being returned by inpcb lookups.
	*
	* It is used by TCP to mark an inpcb as unused and avoid future packet
	* delivery or event notification when a socket remains open but TCP has
	* closed. This might occur as a result of a shutdown()-initiated TCP close
	* or a RST on the wire, and allows the port binding to be reused while still
	* maintaining the invariant that so_pcb always points to a valid inpcb until
	* in_pcbdetach().
	*
	* XXXRW: Possibly in_pcbdrop() should also prevent future notifications by
	* in_pcbnotifyall() and in_pcbpurgeif0()?
	*/
	void
	in_pcbdrop(struct inpcb *inp)
	{

	INP_INFO_WLOCK_ASSERT(inp->inp_pcbinfo);
	INP_WLOCK_ASSERT(inp);

	inp->inp_flags \|= INP_DROPPED;
	if (inp->inp_flags & INP_INHASHLIST) {
	struct inpcbport *phd = inp->inp_phd;

	LIST_REMOVE(inp, inp_hash);
	LIST_REMOVE(inp, inp_portlist);
	if (LIST_FIRST(&phd->phd_pcblist) == NULL) {
	LIST_REMOVE(phd, phd_hash);
	free(phd, M_PCB);
	}
	inp->inp_flags &= ~INP_INHASHLIST;
	}
	}

	#ifdef INET
	/*
	* Common routines to return the socket addresses associated with inpcbs.
	*/
	struct sockaddr *
	in_sockaddr(in_port_t port, struct in_addr *addr_p)
	{
	struct sockaddr_in *sin;

	sin = malloc(sizeof *sin, M_SONAME,
	M_WAITOK \| M_ZERO);
	sin->sin_family = AF_INET;
	sin->sin_len = sizeof(*sin);
	sin->sin_addr = *addr_p;
	sin->sin_port = port;

	return (struct sockaddr *)sin;
	}

	int
	in_getsockaddr(struct socket so, struct sockaddr *nam)
	{
	struct inpcb *inp;
	struct in_addr addr;
	in_port_t port;

	inp = sotoinpcb(so);
	KASSERT(inp != NULL, ("in_getsockaddr: inp == NULL"));

	INP_RLOCK(inp);
	port = inp->inp_lport;
	addr = inp->inp_laddr;
	INP_RUNLOCK(inp);

	*nam = in_sockaddr(port, &addr);
	return 0;
	}

	int
	in_getpeeraddr(struct socket so, struct sockaddr *nam)
	{
	struct inpcb *inp;
	struct in_addr addr;
	in_port_t port;

	inp = sotoinpcb(so);
	KASSERT(inp != NULL, ("in_getpeeraddr: inp == NULL"));

	INP_RLOCK(inp);
	port = inp->inp_fport;
	addr = inp->inp_faddr;
	INP_RUNLOCK(inp);

	*nam = in_sockaddr(port, &addr);
	return 0;
	}

	void
	in_pcbnotifyall(struct inpcbinfo *pcbinfo, struct in_addr faddr, int errno,
	struct inpcb (notify)(struct inpcb *, int))
	{
	struct inpcb inp, inp_temp;

	INP_INFO_WLOCK(pcbinfo);
	LIST_FOREACH_SAFE(inp, pcbinfo->ipi_listhead, inp_list, inp_temp) {
	INP_WLOCK(inp);
	#ifdef INET6
	if ((inp->inp_vflag & INP_IPV4) == 0) {
	INP_WUNLOCK(inp);
	continue;
	}
	#endif
	if (inp->inp_faddr.s_addr != faddr.s_addr \|\|
	inp->inp_socket == NULL) {
	INP_WUNLOCK(inp);
	continue;
	}
	if ((*notify)(inp, errno))
	INP_WUNLOCK(inp);
	}
	INP_INFO_WUNLOCK(pcbinfo);
	}

	void
	in_pcbpurgeif0(struct inpcbinfo pcbinfo, struct ifnet ifp)
	{
	struct inpcb *inp;
	struct ip_moptions *imo;
	int i, gap;

	INP_INFO_RLOCK(pcbinfo);
	LIST_FOREACH(inp, pcbinfo->ipi_listhead, inp_list) {
	INP_WLOCK(inp);
	imo = inp->inp_moptions;
	if ((inp->inp_vflag & INP_IPV4) &&
	imo != NULL) {
	/*
	* Unselect the outgoing interface if it is being
	* detached.
	*/
	if (imo->imo_multicast_ifp == ifp)
	imo->imo_multicast_ifp = NULL;

	/*
	* Drop multicast group membership if we joined
	* through the interface being detached.
	*/
	for (i = 0, gap = 0; i < imo->imo_num_memberships;
	i++) {
	if (imo->imo_membership[i]->inm_ifp == ifp) {
	in_delmulti(imo->imo_membership[i]);
	gap++;
	} else if (gap != 0)
	imo->imo_membership[i - gap] =
	imo->imo_membership[i];
	}
	imo->imo_num_memberships -= gap;
	}
	INP_WUNLOCK(inp);
	}
	INP_INFO_RUNLOCK(pcbinfo);
	}

	/*
	* Lookup a PCB based on the local address and port.
	*/
	#define INP_LOOKUP_MAPPED_PCB_COST 3
	struct inpcb *
	in_pcblookup_local(struct inpcbinfo *pcbinfo, struct in_addr laddr,
	u_short lport, int lookupflags, struct ucred *cred)
	{
	struct inpcb *inp;
	#ifdef INET6
	int matchwild = 3 + INP_LOOKUP_MAPPED_PCB_COST;
	#else
	int matchwild = 3;
	#endif
	int wildcard;

	KASSERT((lookupflags & ~(INPLOOKUP_WILDCARD)) == 0,
	("%s: invalid lookup flags %d", __func__, lookupflags));

	INP_INFO_LOCK_ASSERT(pcbinfo);

	if ((lookupflags & INPLOOKUP_WILDCARD) == 0) {
	struct inpcbhead *head;
	/*
	* Look for an unconnected (wildcard foreign addr) PCB that
	* matches the local address and port we're looking for.
	*/
	head = &pcbinfo->ipi_hashbase[INP_PCBHASH(INADDR_ANY, lport,
	0, pcbinfo->ipi_hashmask)];
	LIST_FOREACH(inp, head, inp_hash) {
	#ifdef INET6
	/* XXX inp locking */
	if ((inp->inp_vflag & INP_IPV4) == 0)
	continue;
	#endif
	if (inp->inp_faddr.s_addr == INADDR_ANY &&
	inp->inp_laddr.s_addr == laddr.s_addr &&
	inp->inp_lport == lport) {
	/*
	* Found?
	*/
	if (cred == NULL \|\|
	prison_equal_ip4(cred->cr_prison,
	inp->inp_cred->cr_prison))
	return (inp);
	}
	}
	/*
	* Not found.
	*/
	return (NULL);
	} else {
	struct inpcbporthead *porthash;
	struct inpcbport *phd;
	struct inpcb *match = NULL;
	/*
	* Best fit PCB lookup.
	*
	* First see if this local port is in use by looking on the
	* port hash list.
	*/
	porthash = &pcbinfo->ipi_porthashbase[INP_PCBPORTHASH(lport,
	pcbinfo->ipi_porthashmask)];
	LIST_FOREACH(phd, porthash, phd_hash) {
	if (phd->phd_port == lport)
	break;
	}
	if (phd != NULL) {
	/*
	* Port is in use by one or more PCBs. Look for best
	* fit.
	*/
	LIST_FOREACH(inp, &phd->phd_pcblist, inp_portlist) {
	wildcard = 0;
	if (cred != NULL &&
	!prison_equal_ip4(inp->inp_cred->cr_prison,
	cred->cr_prison))
	continue;
	#ifdef INET6
	/* XXX inp locking */
	if ((inp->inp_vflag & INP_IPV4) == 0)
	continue;
	/*
	* We never select the PCB that has
	* INP_IPV6 flag and is bound to :: if
	* we have another PCB which is bound
	* to 0.0.0.0. If a PCB has the
	* INP_IPV6 flag, then we set its cost
	* higher than IPv4 only PCBs.
	*
	* Note that the case only happens
	* when a socket is bound to ::, under
	* the condition that the use of the
	* mapped address is allowed.
	*/
	if ((inp->inp_vflag & INP_IPV6) != 0)
	wildcard += INP_LOOKUP_MAPPED_PCB_COST;
	#endif
	if (inp->inp_faddr.s_addr != INADDR_ANY)
	wildcard++;
	if (inp->inp_laddr.s_addr != INADDR_ANY) {
	if (laddr.s_addr == INADDR_ANY)
	wildcard++;
	else if (inp->inp_laddr.s_addr != laddr.s_addr)
	continue;
	} else {
	if (laddr.s_addr != INADDR_ANY)
	wildcard++;
	}
	if (wildcard < matchwild) {
	match = inp;
	matchwild = wildcard;
	if (matchwild == 0)
	break;
	}
	}
	}
	return (match);
	}
	}
	#undef INP_LOOKUP_MAPPED_PCB_COST

	/*
	* Lookup PCB in hash list.
	*/
	struct inpcb *
	in_pcblookup_hash(struct inpcbinfo *pcbinfo, struct in_addr faddr,
	u_int fport_arg, struct in_addr laddr, u_int lport_arg, int lookupflags,
	struct ifnet *ifp)
	{
	struct inpcbhead *head;
	struct inpcb inp, tmpinp;
	u_short fport = fport_arg, lport = lport_arg;

	KASSERT((lookupflags & ~(INPLOOKUP_WILDCARD)) == 0,
	("%s: invalid lookup flags %d", __func__, lookupflags));

	INP_INFO_LOCK_ASSERT(pcbinfo);

	/*
	* First look for an exact match.
	*/
	tmpinp = NULL;
	head = &pcbinfo->ipi_hashbase[INP_PCBHASH(faddr.s_addr, lport, fport,
	pcbinfo->ipi_hashmask)];
	LIST_FOREACH(inp, head, inp_hash) {
	#ifdef INET6
	/* XXX inp locking */
	if ((inp->inp_vflag & INP_IPV4) == 0)
	continue;
	#endif
	if (inp->inp_faddr.s_addr == faddr.s_addr &&
	inp->inp_laddr.s_addr == laddr.s_addr &&
	inp->inp_fport == fport &&
	inp->inp_lport == lport) {
	/*
	* XXX We should be able to directly return
	* the inp here, without any checks.
	* Well unless both bound with SO_REUSEPORT?
	*/
	if (prison_flag(inp->inp_cred, PR_IP4))
	return (inp);
	if (tmpinp == NULL)
	tmpinp = inp;
	}
	}
	if (tmpinp != NULL)
	return (tmpinp);

	/*
	* Then look for a wildcard match, if requested.
	*/
	if ((lookupflags & INPLOOKUP_WILDCARD) != 0) {
	struct inpcb local_wild = NULL, local_exact = NULL;
	#ifdef INET6
	struct inpcb *local_wild_mapped = NULL;
	#endif
	struct inpcb *jail_wild = NULL;
	int injail;

	/*
	* Order of socket selection - we always prefer jails.
	* 1. jailed, non-wild.
	* 2. jailed, wild.
	* 3. non-jailed, non-wild.
	* 4. non-jailed, wild.
	*/

	head = &pcbinfo->ipi_hashbase[INP_PCBHASH(INADDR_ANY, lport,
	0, pcbinfo->ipi_hashmask)];
	LIST_FOREACH(inp, head, inp_hash) {
	#ifdef INET6
	/* XXX inp locking */
	if ((inp->inp_vflag & INP_IPV4) == 0)
	continue;
	#endif
	if (inp->inp_faddr.s_addr != INADDR_ANY \|\|
	inp->inp_lport != lport)
	continue;

	/* XXX inp locking */
	if (ifp && ifp->if_type == IFT_FAITH &&
	(inp->inp_flags & INP_FAITH) == 0)
	continue;

	injail = prison_flag(inp->inp_cred, PR_IP4);
	if (injail) {
	if (prison_check_ip4(inp->inp_cred,
	&laddr) != 0)
	continue;
	} else {
	if (local_exact != NULL)
	continue;
	}

	if (inp->inp_laddr.s_addr == laddr.s_addr) {
	if (injail)
	return (inp);
	else
	local_exact = inp;
	} else if (inp->inp_laddr.s_addr == INADDR_ANY) {
	#ifdef INET6
	/* XXX inp locking, NULL check */
	if (inp->inp_vflag & INP_IPV6PROTO)
	local_wild_mapped = inp;
	else
	#endif /* INET6 */
	if (injail)
	jail_wild = inp;
	else
	local_wild = inp;
	}
	} /* LIST_FOREACH */
	if (jail_wild != NULL)
	return (jail_wild);
	if (local_exact != NULL)
	return (local_exact);
	if (local_wild != NULL)
	return (local_wild);
	#ifdef INET6
	if (local_wild_mapped != NULL)
	return (local_wild_mapped);
	#endif /* defined(INET6) */
	} /* if ((lookupflags & INPLOOKUP_WILDCARD) != 0) */

	return (NULL);
	}
	#endif /* INET */

	/*
	* Insert PCB onto various hash lists.
	*/
	int
	in_pcbinshash(struct inpcb *inp)
	{
	struct inpcbhead *pcbhash;
	struct inpcbporthead *pcbporthash;
	struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;
	struct inpcbport *phd;
	u_int32_t hashkey_faddr;

	INP_INFO_WLOCK_ASSERT(pcbinfo);
	INP_WLOCK_ASSERT(inp);
	KASSERT((inp->inp_flags & INP_INHASHLIST) == 0,
	("in_pcbinshash: INP_INHASHLIST"));

	#ifdef INET6
	if (inp->inp_vflag & INP_IPV6)
	hashkey_faddr = inp->in6p_faddr.s6_addr32[3] /* XXX */;
	else
	#endif /* INET6 */
	hashkey_faddr = inp->inp_faddr.s_addr;

	pcbhash = &pcbinfo->ipi_hashbase[INP_PCBHASH(hashkey_faddr,
	inp->inp_lport, inp->inp_fport, pcbinfo->ipi_hashmask)];

	pcbporthash = &pcbinfo->ipi_porthashbase[
	INP_PCBPORTHASH(inp->inp_lport, pcbinfo->ipi_porthashmask)];

	/*
	* Go through port list and look for a head for this lport.
	*/
	LIST_FOREACH(phd, pcbporthash, phd_hash) {
	if (phd->phd_port == inp->inp_lport)
	break;
	}
	/*
	* If none exists, malloc one and tack it on.
	*/
	if (phd == NULL) {
	phd = malloc(sizeof(struct inpcbport), M_PCB, M_NOWAIT);
	if (phd == NULL) {
	return (ENOBUFS); /* XXX */
	}
	phd->phd_port = inp->inp_lport;
	LIST_INIT(&phd->phd_pcblist);
	LIST_INSERT_HEAD(pcbporthash, phd, phd_hash);
	}
	inp->inp_phd = phd;
	LIST_INSERT_HEAD(&phd->phd_pcblist, inp, inp_portlist);
	LIST_INSERT_HEAD(pcbhash, inp, inp_hash);
	inp->inp_flags \|= INP_INHASHLIST;
	return (0);
	}

	/*
	* Move PCB to the proper hash bucket when { faddr, fport } have been
	* changed. NOTE: This does not handle the case of the lport changing (the
	* hashed port list would have to be updated as well), so the lport must
	* not change after in_pcbinshash() has been called.
	*/
	void
	in_pcbrehash(struct inpcb *inp)
	{
	struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;
	struct inpcbhead *head;
	u_int32_t hashkey_faddr;

	INP_INFO_WLOCK_ASSERT(pcbinfo);
	INP_WLOCK_ASSERT(inp);
	KASSERT(inp->inp_flags & INP_INHASHLIST,
	("in_pcbrehash: !INP_INHASHLIST"));

	#ifdef INET6
	if (inp->inp_vflag & INP_IPV6)
	hashkey_faddr = inp->in6p_faddr.s6_addr32[3] /* XXX */;
	else
	#endif /* INET6 */
	hashkey_faddr = inp->inp_faddr.s_addr;

	head = &pcbinfo->ipi_hashbase[INP_PCBHASH(hashkey_faddr,
	inp->inp_lport, inp->inp_fport, pcbinfo->ipi_hashmask)];

	LIST_REMOVE(inp, inp_hash);
	LIST_INSERT_HEAD(head, inp, inp_hash);
	}

	/*
	* Remove PCB from various lists.
	*/
	static void
	in_pcbremlists(struct inpcb *inp)
	{
	struct inpcbinfo *pcbinfo = inp->inp_pcbinfo;

	INP_INFO_WLOCK_ASSERT(pcbinfo);
	INP_WLOCK_ASSERT(inp);

	inp->inp_gencnt = ++pcbinfo->ipi_gencnt;
	if (inp->inp_flags & INP_INHASHLIST) {
	struct inpcbport *phd = inp->inp_phd;

	LIST_REMOVE(inp, inp_hash);
	LIST_REMOVE(inp, inp_portlist);
	if (LIST_FIRST(&phd->phd_pcblist) == NULL) {
	LIST_REMOVE(phd, phd_hash);
	free(phd, M_PCB);
	}
	inp->inp_flags &= ~INP_INHASHLIST;
	}
	LIST_REMOVE(inp, inp_list);
	pcbinfo->ipi_count--;
	}

	/*
	* A set label operation has occurred at the socket layer, propagate the
	* label change into the in_pcb for the socket.
	*/
	void
	in_pcbsosetlabel(struct socket *so)
	{
	#ifdef MAC
	struct inpcb *inp;

	inp = sotoinpcb(so);
	KASSERT(inp != NULL, ("in_pcbsosetlabel: so->so_pcb == NULL"));

	INP_WLOCK(inp);
	SOCK_LOCK(so);
	mac_inpcb_sosetlabel(so, inp);
	SOCK_UNLOCK(so);
	INP_WUNLOCK(inp);
	#endif
	}

	/*
	* ipport_tick runs once per second, determining if random port allocation
	* should be continued. If more than ipport_randomcps ports have been
	* allocated in the last second, then we return to sequential port
	* allocation. We return to random allocation only once we drop below
	* ipport_randomcps for at least ipport_randomtime seconds.
	*/
	static void
	ipport_tick(void *xtp)
	{
	VNET_ITERATOR_DECL(vnet_iter);

	VNET_LIST_RLOCK_NOSLEEP();
	VNET_FOREACH(vnet_iter) {
	CURVNET_SET(vnet_iter); /* XXX appease INVARIANTS here */
	if (V_ipport_tcpallocs <=
	V_ipport_tcplastcount + V_ipport_randomcps) {
	if (V_ipport_stoprandom > 0)
	V_ipport_stoprandom--;
	} else
	V_ipport_stoprandom = V_ipport_randomtime;
	V_ipport_tcplastcount = V_ipport_tcpallocs;
	CURVNET_RESTORE();
	}
	VNET_LIST_RUNLOCK_NOSLEEP();
	callout_reset(&ipport_tick_callout, hz, ipport_tick, NULL);
	}

	static void
	ip_fini(void *xtp)
	{

	callout_stop(&ipport_tick_callout);
	}

	/*
	* The ipport_callout should start running at about the time we attach the
	* inet or inet6 domains.
	*/
	static void
	ipport_tick_init(const void *unused __unused)
	{

	/* Start ipport_tick. */
	callout_init(&ipport_tick_callout, CALLOUT_MPSAFE);
	callout_reset(&ipport_tick_callout, 1, ipport_tick, NULL);
	EVENTHANDLER_REGISTER(shutdown_pre_sync, ip_fini, NULL,
	SHUTDOWN_PRI_DEFAULT);
	}
	SYSINIT(ipport_tick_init, SI_SUB_PROTO_DOMAIN, SI_ORDER_MIDDLE,
	ipport_tick_init, NULL);

	void
	inp_wlock(struct inpcb *inp)
	{

	INP_WLOCK(inp);
	}

	void
	inp_wunlock(struct inpcb *inp)
	{

	INP_WUNLOCK(inp);
	}

	void
	inp_rlock(struct inpcb *inp)
	{

	INP_RLOCK(inp);
	}

	void
	inp_runlock(struct inpcb *inp)
	{

	INP_RUNLOCK(inp);
	}

	#ifdef INVARIANTS
	void
	inp_lock_assert(struct inpcb *inp)
	{

	INP_WLOCK_ASSERT(inp);
	}

	void
	inp_unlock_assert(struct inpcb *inp)
	{

	INP_UNLOCK_ASSERT(inp);
	}
	#endif

	void
	inp_apply_all(void (func)(struct inpcb , void ), void arg)
	{
	struct inpcb *inp;

	INP_INFO_RLOCK(&V_tcbinfo);
	LIST_FOREACH(inp, V_tcbinfo.ipi_listhead, inp_list) {
	INP_WLOCK(inp);
	func(inp, arg);
	INP_WUNLOCK(inp);
	}
	INP_INFO_RUNLOCK(&V_tcbinfo);
	}

	struct socket *
	inp_inpcbtosocket(struct inpcb *inp)
	{

	INP_WLOCK_ASSERT(inp);
	return (inp->inp_socket);
	}

	struct tcpcb *
	inp_inpcbtotcpcb(struct inpcb *inp)
	{

	INP_WLOCK_ASSERT(inp);
	return ((struct tcpcb *)inp->inp_ppcb);
	}

	int
	inp_ip_tos_get(const struct inpcb *inp)
	{

	return (inp->inp_ip_tos);
	}

	void
	inp_ip_tos_set(struct inpcb *inp, int val)
	{

	inp->inp_ip_tos = val;
	}

	void
	inp_4tuple_get(struct inpcb inp, uint32_t laddr, uint16_t *lp,
	uint32_t faddr, uint16_t fp)
	{

	INP_LOCK_ASSERT(inp);
	*laddr = inp->inp_laddr.s_addr;
	*faddr = inp->inp_faddr.s_addr;
	*lp = inp->inp_lport;
	*fp = inp->inp_fport;
	}

	struct inpcb *
	so_sotoinpcb(struct socket *so)
	{

	return (sotoinpcb(so));
	}

	struct tcpcb *
	so_sototcpcb(struct socket *so)
	{

	return (sototcpcb(so));
	}

	#ifdef DDB
	static void
	db_print_indent(int indent)
	{
	int i;

	for (i = 0; i < indent; i++)
	db_printf(" ");
	}

	static void
	db_print_inconninfo(struct in_conninfo inc, const char name, int indent)
	{
	char faddr_str[48], laddr_str[48];

	db_print_indent(indent);
	db_printf("%s at %p\n", name, inc);

	indent += 2;

	#ifdef INET6
	if (inc->inc_flags & INC_ISIPV6) {
	/* IPv6. */
	ip6_sprintf(laddr_str, &inc->inc6_laddr);
	ip6_sprintf(faddr_str, &inc->inc6_faddr);
	} else {
	#endif
	/* IPv4. */
	inet_ntoa_r(inc->inc_laddr, laddr_str);
	inet_ntoa_r(inc->inc_faddr, faddr_str);
	#ifdef INET6
	}
	#endif
	db_print_indent(indent);
	db_printf("inc_laddr %s inc_lport %u\n", laddr_str,
	ntohs(inc->inc_lport));
	db_print_indent(indent);
	db_printf("inc_faddr %s inc_fport %u\n", faddr_str,
	ntohs(inc->inc_fport));
	}

	static void
	db_print_inpflags(int inp_flags)
	{
	int comma;

	comma = 0;
	if (inp_flags & INP_RECVOPTS) {
	db_printf("%sINP_RECVOPTS", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_RECVRETOPTS) {
	db_printf("%sINP_RECVRETOPTS", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_RECVDSTADDR) {
	db_printf("%sINP_RECVDSTADDR", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_HDRINCL) {
	db_printf("%sINP_HDRINCL", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_HIGHPORT) {
	db_printf("%sINP_HIGHPORT", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_LOWPORT) {
	db_printf("%sINP_LOWPORT", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_ANONPORT) {
	db_printf("%sINP_ANONPORT", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_RECVIF) {
	db_printf("%sINP_RECVIF", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_MTUDISC) {
	db_printf("%sINP_MTUDISC", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_FAITH) {
	db_printf("%sINP_FAITH", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_RECVTTL) {
	db_printf("%sINP_RECVTTL", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_DONTFRAG) {
	db_printf("%sINP_DONTFRAG", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_IPV6_V6ONLY) {
	db_printf("%sIN6P_IPV6_V6ONLY", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_PKTINFO) {
	db_printf("%sIN6P_PKTINFO", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_HOPLIMIT) {
	db_printf("%sIN6P_HOPLIMIT", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_HOPOPTS) {
	db_printf("%sIN6P_HOPOPTS", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_DSTOPTS) {
	db_printf("%sIN6P_DSTOPTS", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_RTHDR) {
	db_printf("%sIN6P_RTHDR", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_RTHDRDSTOPTS) {
	db_printf("%sIN6P_RTHDRDSTOPTS", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_TCLASS) {
	db_printf("%sIN6P_TCLASS", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_AUTOFLOWLABEL) {
	db_printf("%sIN6P_AUTOFLOWLABEL", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_TIMEWAIT) {
	db_printf("%sINP_TIMEWAIT", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_ONESBCAST) {
	db_printf("%sINP_ONESBCAST", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_DROPPED) {
	db_printf("%sINP_DROPPED", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & INP_SOCKREF) {
	db_printf("%sINP_SOCKREF", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_RFC2292) {
	db_printf("%sIN6P_RFC2292", comma ? ", " : "");
	comma = 1;
	}
	if (inp_flags & IN6P_MTU) {
	db_printf("IN6P_MTU%s", comma ? ", " : "");
	comma = 1;
	}
	}

	static void
	db_print_inpvflag(u_char inp_vflag)
	{
	int comma;

	comma = 0;
	if (inp_vflag & INP_IPV4) {
	db_printf("%sINP_IPV4", comma ? ", " : "");
	comma = 1;
	}
	if (inp_vflag & INP_IPV6) {
	db_printf("%sINP_IPV6", comma ? ", " : "");
	comma = 1;
	}
	if (inp_vflag & INP_IPV6PROTO) {
	db_printf("%sINP_IPV6PROTO", comma ? ", " : "");
	comma = 1;
	}
	}

	static void
	db_print_inpcb(struct inpcb inp, const char name, int indent)
	{

	db_print_indent(indent);
	db_printf("%s at %p\n", name, inp);

	indent += 2;

	db_print_indent(indent);
	db_printf("inp_flow: 0x%x\n", inp->inp_flow);

	db_print_inconninfo(&inp->inp_inc, "inp_conninfo", indent);

	db_print_indent(indent);
	db_printf("inp_ppcb: %p inp_pcbinfo: %p inp_socket: %p\n",
	inp->inp_ppcb, inp->inp_pcbinfo, inp->inp_socket);

	db_print_indent(indent);
	db_printf("inp_label: %p inp_flags: 0x%x (",
	inp->inp_label, inp->inp_flags);
	db_print_inpflags(inp->inp_flags);
	db_printf(")\n");

	db_print_indent(indent);
	db_printf("inp_sp: %p inp_vflag: 0x%x (", inp->inp_sp,
	inp->inp_vflag);
	db_print_inpvflag(inp->inp_vflag);
	db_printf(")\n");

	db_print_indent(indent);
	db_printf("inp_ip_ttl: %d inp_ip_p: %d inp_ip_minttl: %d\n",
	inp->inp_ip_ttl, inp->inp_ip_p, inp->inp_ip_minttl);

	db_print_indent(indent);
	#ifdef INET6
	if (inp->inp_vflag & INP_IPV6) {
	db_printf("in6p_options: %p in6p_outputopts: %p "
	"in6p_moptions: %p\n", inp->in6p_options,
	inp->in6p_outputopts, inp->in6p_moptions);
	db_printf("in6p_icmp6filt: %p in6p_cksum %d "
	"in6p_hops %u\n", inp->in6p_icmp6filt, inp->in6p_cksum,
	inp->in6p_hops);
	} else
	#endif
	{
	db_printf("inp_ip_tos: %d inp_ip_options: %p "
	"inp_ip_moptions: %p\n", inp->inp_ip_tos,
	inp->inp_options, inp->inp_moptions);
	}

	db_print_indent(indent);
	db_printf("inp_phd: %p inp_gencnt: %ju\n", inp->inp_phd,
	(uintmax_t)inp->inp_gencnt);
	}

	DB_SHOW_COMMAND(inpcb, db_show_inpcb)
	{
	struct inpcb *inp;

	if (!have_addr) {
	db_printf("usage: show inpcb <addr>\n");
	return;
	}
	inp = (struct inpcb *)addr;

	db_print_inpcb(inp, "inpcb", 0);
	}
	#endif
	diff --git a/sys/netinet/in_pcb.h b/sys/netinet/in_pcb.h
	index 52debb063ce2..14d4ea233751 100644
	--- a/sys/netinet/in_pcb.h
	+++ b/sys/netinet/in_pcb.h
	@@ -1,545 +1,547 @@
	/*-
	* Copyright (c) 1982, 1986, 1990, 1993
	* The Regents of the University of California.
	* Copyright (c) 2010-2011 Juniper Networks, Inc.
	* All rights reserved.
	*
	* Portions of this software were developed by Robert N. M. Watson under
	* contract to Juniper Networks, Inc.
	*
	* 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.
	*
	* @(#)in_pcb.h 8.1 (Berkeley) 6/10/93
	* $FreeBSD$
	*/

	#ifndef _NETINET_IN_PCB_H_
	#define _NETINET_IN_PCB_H_

	#include <sys/queue.h>
	#include <sys/_lock.h>
	#include <sys/_mutex.h>
	#include <sys/_rwlock.h>

	#ifdef _KERNEL
	#include <sys/rwlock.h>
	#include <net/vnet.h>
	#include <vm/uma.h>
	#endif

	#define in6pcb inpcb /* for KAME src sync over BSD's /
	#define in6p_sp inp_sp /* for KAME src sync over BSD's /
	struct inpcbpolicy;

	/*
	* struct inpcb is the common protocol control block structure used in most
	* IP transport protocols.
	*
	* Pointers to local and foreign host table entries, local and foreign socket
	* numbers, and pointers up (to a socket structure) and down (to a
	* protocol-specific control block) are stored here.
	*/
	LIST_HEAD(inpcbhead, inpcb);
	LIST_HEAD(inpcbporthead, inpcbport);
	typedef u_quad_t inp_gen_t;

	/*
	* PCB with AF_INET6 null bind'ed laddr can receive AF_INET input packet.
	* So, AF_INET6 null laddr is also used as AF_INET null laddr, by utilizing
	* the following structure.
	*/
	struct in_addr_4in6 {
	u_int32_t ia46_pad32[3];
	struct in_addr ia46_addr4;
	};

	/*
	* NOTE: ipv6 addrs should be 64-bit aligned, per RFC 2553. in_conninfo has
	* some extra padding to accomplish this.
	*/
	struct in_endpoints {
	u_int16_t ie_fport; /* foreign port */
	u_int16_t ie_lport; /* local port */
	/* protocol dependent part, local and foreign addr */
	union {
	/* foreign host table entry */
	struct in_addr_4in6 ie46_foreign;
	struct in6_addr ie6_foreign;
	} ie_dependfaddr;
	union {
	/* local host table entry */
	struct in_addr_4in6 ie46_local;
	struct in6_addr ie6_local;
	} ie_dependladdr;
	};
	#define ie_faddr ie_dependfaddr.ie46_foreign.ia46_addr4
	#define ie_laddr ie_dependladdr.ie46_local.ia46_addr4
	#define ie6_faddr ie_dependfaddr.ie6_foreign
	#define ie6_laddr ie_dependladdr.ie6_local

	/*
	* XXX The defines for inc_* are hacks and should be changed to direct
	* references.
	*/
	struct in_conninfo {
	u_int8_t inc_flags;
	u_int8_t inc_len;
	u_int16_t inc_fibnum; /* XXX was pad, 16 bits is plenty */
	/* protocol dependent part */
	struct in_endpoints inc_ie;
	};

	/*
	* Flags for inc_flags.
	*/
	#define INC_ISIPV6 0x01

	#define inc_isipv6 inc_flags /* temp compatability */
	#define inc_fport inc_ie.ie_fport
	#define inc_lport inc_ie.ie_lport
	#define inc_faddr inc_ie.ie_faddr
	#define inc_laddr inc_ie.ie_laddr
	#define inc6_faddr inc_ie.ie6_faddr
	#define inc6_laddr inc_ie.ie6_laddr

	struct icmp6_filter;

	/*-
	* struct inpcb captures the network layer state for TCP, UDP, and raw IPv4
	* and IPv6 sockets. In the case of TCP, further per-connection state is
	* hung off of inp_ppcb most of the time. Almost all fields of struct inpcb
	* are static after creation or protected by a per-inpcb rwlock, inp_lock. A
	* few fields also require the global pcbinfo lock for the inpcb to be held,
	* when modified, such as the global connection lists and hashes, as well as
	* binding information (which affects which hash a connection is on). This
	* model means that connections can be looked up without holding the
	* per-connection lock, which is important for performance when attempting to
	* find the connection for a packet given its IP and port tuple. Writing to
	* these fields that write locks be held on both the inpcb and global locks.
	*
	* Key:
	* (c) - Constant after initialization
	* (i) - Protected by the inpcb lock
	* (p) - Protected by the pcbinfo lock for the inpcb
	* (s) - Protected by another subsystem's locks
	* (x) - Undefined locking
	*
	* A few other notes:
	*
	* When a read lock is held, stability of the field is guaranteed; to write
	* to a field, a write lock must generally be held.
	*
	* netinet/netinet6-layer code should not assume that the inp_socket pointer
	* is safe to dereference without inp_lock being held, even for protocols
	* other than TCP (where the inpcb persists during TIMEWAIT even after the
	* socket has been freed), or there may be close(2)-related races.
	*
	* The inp_vflag field is overloaded, and would otherwise ideally be (c).
	*/
	struct inpcb {
	LIST_ENTRY(inpcb) inp_hash; /* (i/p) hash list */
	LIST_ENTRY(inpcb) inp_list; /* (i/p) list for all PCBs for proto */
	void inp_ppcb; / (i) pointer to per-protocol pcb */
	struct inpcbinfo inp_pcbinfo; / (c) PCB list info */
	struct socket inp_socket; / (i) back pointer to socket */
	struct ucred inp_cred; / (c) cache of socket cred */
	u_int32_t inp_flow; /* (i) IPv6 flow information */
	int inp_flags; /* (i) generic IP/datagram flags */
	int inp_flags2; /* (i) generic IP/datagram flags #2*/
	u_char inp_vflag; /* (i) IP version flag (v4/v6) */
	u_char inp_ip_ttl; /* (i) time to live proto */
	u_char inp_ip_p; /* (c) protocol proto */
	u_char inp_ip_minttl; /* (i) minimum TTL or drop */
	uint32_t inp_flowid; /* (x) flow id / queue id */
	u_int inp_refcount; /* (i) refcount */
	void inp_pspare[4]; / (x) rtentry / general use */
	u_int inp_ispare[4]; /* general use */

	/* Local and foreign ports, local and foreign addr. */
	struct in_conninfo inp_inc; /* (i/p) list for PCB's local port */

	/* MAC and IPSEC policy information. */
	struct label inp_label; / (i) MAC label */
	struct inpcbpolicy inp_sp; / (s) for IPSEC */

	/* Protocol-dependent part; options. */
	struct {
	u_char inp4_ip_tos; /* (i) type of service proto */
	struct mbuf inp4_options; / (i) IP options */
	struct ip_moptions inp4_moptions; / (i) IP mcast options */
	} inp_depend4;
	struct {
	/* (i) IP options */
	struct mbuf *inp6_options;
	/* (i) IP6 options for outgoing packets */
	struct ip6_pktopts *inp6_outputopts;
	/* (i) IP multicast options */
	struct ip6_moptions *inp6_moptions;
	/* (i) ICMPv6 code type filter */
	struct icmp6_filter *inp6_icmp6filt;
	/* (i) IPV6_CHECKSUM setsockopt */
	int inp6_cksum;
	short inp6_hops;
	} inp_depend6;
	LIST_ENTRY(inpcb) inp_portlist; /* (i/p) */
	struct inpcbport inp_phd; / (i/p) head of this list */
	#define inp_zero_size offsetof(struct inpcb, inp_gencnt)
	inp_gen_t inp_gencnt; /* (c) generation count */
	struct llentry inp_lle; / cached L2 information */
	struct rtentry inp_rt; / cached L3 information */
	struct rwlock inp_lock;
	};
	#define inp_fport inp_inc.inc_fport
	#define inp_lport inp_inc.inc_lport
	#define inp_faddr inp_inc.inc_faddr
	#define inp_laddr inp_inc.inc_laddr
	#define inp_ip_tos inp_depend4.inp4_ip_tos
	#define inp_options inp_depend4.inp4_options
	#define inp_moptions inp_depend4.inp4_moptions

	#define in6p_faddr inp_inc.inc6_faddr
	#define in6p_laddr inp_inc.inc6_laddr
	#define in6p_hops inp_depend6.inp6_hops /* default hop limit */
	#define in6p_flowinfo inp_flow
	#define in6p_options inp_depend6.inp6_options
	#define in6p_outputopts inp_depend6.inp6_outputopts
	#define in6p_moptions inp_depend6.inp6_moptions
	#define in6p_icmp6filt inp_depend6.inp6_icmp6filt
	#define in6p_cksum inp_depend6.inp6_cksum

	#define inp_vnet inp_pcbinfo->ipi_vnet

	/*
	* The range of the generation count, as used in this implementation, is 9e19.
	* We would have to create 300 billion connections per second for this number
	* to roll over in a year. This seems sufficiently unlikely that we simply
	* don't concern ourselves with that possibility.
	*/

	/*
	* Interface exported to userland by various protocols which use inpcbs. Hack
	* alert -- only define if struct xsocket is in scope.
	*/
	#ifdef _SYS_SOCKETVAR_H_
	struct xinpcb {
	size_t xi_len; /* length of this structure */
	struct inpcb xi_inp;
	struct xsocket xi_socket;
	u_quad_t xi_alignment_hack;
	};

	struct xinpgen {
	size_t xig_len; /* length of this structure */
	u_int xig_count; /* number of PCBs at this time */
	inp_gen_t xig_gen; /* generation count at this time */
	so_gen_t xig_sogen; /* socket generation count at this time */
	};
	#endif /* _SYS_SOCKETVAR_H_ */

	struct inpcbport {
	LIST_ENTRY(inpcbport) phd_hash;
	struct inpcbhead phd_pcblist;
	u_short phd_port;
	};

	/*-
	* Global data structure for each high-level protocol (UDP, TCP, ...) in both
	* IPv4 and IPv6. Holds inpcb lists and information for managing them.
	*
	* Each pcbinfo is protected by ipi_lock, covering mutable global fields (such
	* as the global pcb list) and hashed lookup tables. The lock order is:
	*
	* ipi_lock (before) inpcb locks
	*
	* Locking key:
	*
	* (c) Constant or nearly constant after initialisation
	* (g) Locked by ipi_lock
	* (h) Read using either ipi_lock or inpcb lock; write requires both.
	* (x) Synchronisation properties poorly defined
	*/
	struct inpcbinfo {
	/*
	* Global lock protecting global inpcb list, inpcb count, hash tables,
	* etc.
	*/
	struct rwlock ipi_lock;

	/*
	* Global list of inpcbs on the protocol.
	*/
	struct inpcbhead ipi_listhead; / (g) */
	u_int ipi_count; /* (g) */

	/*
	* Generation count -- incremented each time a connection is allocated
	* or freed.
	*/
	u_quad_t ipi_gencnt; /* (g) */

	/*
	* Fields associated with port lookup and allocation.
	*/
	u_short ipi_lastport; /* (x) */
	u_short ipi_lastlow; /* (x) */
	u_short ipi_lasthi; /* (x) */

	/*
	* UMA zone from which inpcbs are allocated for this protocol.
	*/
	struct uma_zone ipi_zone; / (c) */

	/*
	* Global hash of inpcbs, hashed by local and foreign addresses and
	* port numbers.
	*/
	struct inpcbhead ipi_hashbase; / (g) */
	u_long ipi_hashmask; /* (g) */

	/*
	* Global hash of inpcbs, hashed by only local port number.
	*/
	struct inpcbporthead ipi_porthashbase; / (g) */
	u_long ipi_porthashmask; /* (g) */

	/*
	* Pointer to network stack instance
	*/
	struct vnet ipi_vnet; / (c) */

	/*
	* general use 2
	*/
	void *ipi_pspare[2];
	};

	#define INP_LOCK_INIT(inp, d, t) \
	rw_init_flags(&(inp)->inp_lock, (t), RW_RECURSE \| RW_DUPOK)
	#define INP_LOCK_DESTROY(inp) rw_destroy(&(inp)->inp_lock)
	#define INP_RLOCK(inp) rw_rlock(&(inp)->inp_lock)
	#define INP_WLOCK(inp) rw_wlock(&(inp)->inp_lock)
	#define INP_TRY_RLOCK(inp) rw_try_rlock(&(inp)->inp_lock)
	#define INP_TRY_WLOCK(inp) rw_try_wlock(&(inp)->inp_lock)
	#define INP_RUNLOCK(inp) rw_runlock(&(inp)->inp_lock)
	#define INP_WUNLOCK(inp) rw_wunlock(&(inp)->inp_lock)
	#define INP_TRY_UPGRADE(inp) rw_try_upgrade(&(inp)->inp_lock)
	#define INP_DOWNGRADE(inp) rw_downgrade(&(inp)->inp_lock)
	#define INP_WLOCKED(inp) rw_wowned(&(inp)->inp_lock)
	#define INP_LOCK_ASSERT(inp) rw_assert(&(inp)->inp_lock, RA_LOCKED)
	#define INP_RLOCK_ASSERT(inp) rw_assert(&(inp)->inp_lock, RA_RLOCKED)
	#define INP_WLOCK_ASSERT(inp) rw_assert(&(inp)->inp_lock, RA_WLOCKED)
	#define INP_UNLOCK_ASSERT(inp) rw_assert(&(inp)->inp_lock, RA_UNLOCKED)

	#ifdef _KERNEL
	/*
	* These locking functions are for inpcb consumers outside of sys/netinet,
	* more specifically, they were added for the benefit of TOE drivers. The
	* macros are reserved for use by the stack.
	*/
	void inp_wlock(struct inpcb *);
	void inp_wunlock(struct inpcb *);
	void inp_rlock(struct inpcb *);
	void inp_runlock(struct inpcb *);

	#ifdef INVARIANTS
	void inp_lock_assert(struct inpcb *);
	void inp_unlock_assert(struct inpcb *);
	#else
	static __inline void
	inp_lock_assert(struct inpcb *inp __unused)
	{
	}

	static __inline void
	inp_unlock_assert(struct inpcb *inp __unused)
	{
	}

	#endif

	void inp_apply_all(void (func)(struct inpcb , void ), void arg);
	int inp_ip_tos_get(const struct inpcb *inp);
	void inp_ip_tos_set(struct inpcb *inp, int val);
	struct socket *
	inp_inpcbtosocket(struct inpcb *inp);
	struct tcpcb *
	inp_inpcbtotcpcb(struct inpcb *inp);
	void inp_4tuple_get(struct inpcb inp, uint32_t laddr, uint16_t *lp,
	uint32_t faddr, uint16_t fp);

	#endif /* _KERNEL */

	#define INP_INFO_LOCK_INIT(ipi, d) \
	rw_init_flags(&(ipi)->ipi_lock, (d), RW_RECURSE)
	#define INP_INFO_LOCK_DESTROY(ipi) rw_destroy(&(ipi)->ipi_lock)
	#define INP_INFO_RLOCK(ipi) rw_rlock(&(ipi)->ipi_lock)
	#define INP_INFO_WLOCK(ipi) rw_wlock(&(ipi)->ipi_lock)
	#define INP_INFO_TRY_RLOCK(ipi) rw_try_rlock(&(ipi)->ipi_lock)
	#define INP_INFO_TRY_WLOCK(ipi) rw_try_wlock(&(ipi)->ipi_lock)
	#define INP_INFO_TRY_UPGRADE(ipi) rw_try_upgrade(&(ipi)->ipi_lock)
	#define INP_INFO_RUNLOCK(ipi) rw_runlock(&(ipi)->ipi_lock)
	#define INP_INFO_WUNLOCK(ipi) rw_wunlock(&(ipi)->ipi_lock)
	#define INP_INFO_LOCK_ASSERT(ipi) rw_assert(&(ipi)->ipi_lock, RA_LOCKED)
	#define INP_INFO_RLOCK_ASSERT(ipi) rw_assert(&(ipi)->ipi_lock, RA_RLOCKED)
	#define INP_INFO_WLOCK_ASSERT(ipi) rw_assert(&(ipi)->ipi_lock, RA_WLOCKED)
	#define INP_INFO_UNLOCK_ASSERT(ipi) rw_assert(&(ipi)->ipi_lock, RA_UNLOCKED)

	#define INP_PCBHASH(faddr, lport, fport, mask) \
	(((faddr) ^ ((faddr) >> 16) ^ ntohs((lport) ^ (fport))) & (mask))
	#define INP_PCBPORTHASH(lport, mask) \
	(ntohs((lport)) & (mask))

	/*
	* Flags for inp_vflags -- historically version flags only
	*/
	#define INP_IPV4 0x1
	#define INP_IPV6 0x2
	#define INP_IPV6PROTO 0x4 /* opened under IPv6 protocol */

	/*
	* Flags for inp_flags.
	*/
	#define INP_RECVOPTS 0x00000001 /* receive incoming IP options */
	#define INP_RECVRETOPTS 0x00000002 /* receive IP options for reply */
	#define INP_RECVDSTADDR 0x00000004 /* receive IP dst address */
	#define INP_HDRINCL 0x00000008 /* user supplies entire IP header */
	#define INP_HIGHPORT 0x00000010 /* user wants "high" port binding */
	#define INP_LOWPORT 0x00000020 /* user wants "low" port binding */
	#define INP_ANONPORT 0x00000040 /* port chosen for user */
	#define INP_RECVIF 0x00000080 /* receive incoming interface */
	#define INP_MTUDISC 0x00000100 /* user can do MTU discovery */
	#define INP_FAITH 0x00000200 /* accept FAITH'ed connections */
	#define INP_RECVTTL 0x00000400 /* receive incoming IP TTL */
	#define INP_DONTFRAG 0x00000800 /* don't fragment packet */
	#define INP_BINDANY 0x00001000 /* allow bind to any address */
	#define INP_INHASHLIST 0x00002000 /* in_pcbinshash() has been called */
	#define IN6P_IPV6_V6ONLY 0x00008000 /* restrict AF_INET6 socket for v6 */
	#define IN6P_PKTINFO 0x00010000 /* receive IP6 dst and I/F */
	#define IN6P_HOPLIMIT 0x00020000 /* receive hoplimit */
	#define IN6P_HOPOPTS 0x00040000 /* receive hop-by-hop options */
	#define IN6P_DSTOPTS 0x00080000 /* receive dst options after rthdr */
	#define IN6P_RTHDR 0x00100000 /* receive routing header */
	#define IN6P_RTHDRDSTOPTS 0x00200000 /* receive dstoptions before rthdr */
	#define IN6P_TCLASS 0x00400000 /* receive traffic class value */
	#define IN6P_AUTOFLOWLABEL 0x00800000 /* attach flowlabel automatically */
	#define INP_TIMEWAIT 0x01000000 /* in TIMEWAIT, ppcb is tcptw */
	#define INP_ONESBCAST 0x02000000 /* send all-ones broadcast */
	#define INP_DROPPED 0x04000000 /* protocol drop flag */
	#define INP_SOCKREF 0x08000000 /* strong socket reference */
	#define INP_SW_FLOWID 0x10000000 /* software generated flow id */
	#define INP_HW_FLOWID 0x20000000 /* hardware generated flow id */
	#define IN6P_RFC2292 0x40000000 /* used RFC2292 API on the socket */
	#define IN6P_MTU 0x80000000 /* receive path MTU */

	#define INP_CONTROLOPTS (INP_RECVOPTS\|INP_RECVRETOPTS\|INP_RECVDSTADDR\|\
	INP_RECVIF\|INP_RECVTTL\|\
	IN6P_PKTINFO\|IN6P_HOPLIMIT\|IN6P_HOPOPTS\|\
	IN6P_DSTOPTS\|IN6P_RTHDR\|IN6P_RTHDRDSTOPTS\|\
	IN6P_TCLASS\|IN6P_AUTOFLOWLABEL\|IN6P_RFC2292\|\
	IN6P_MTU)

	/*
	* Flags for inp_flags2.
	*/
	#define INP_LLE_VALID 0x00000001 /* cached lle is valid */
	#define INP_RT_VALID 0x00000002 /* cached rtentry is valid */

	#define INPLOOKUP_WILDCARD 1
	#define sotoinpcb(so) ((struct inpcb *)(so)->so_pcb)
	#define sotoin6pcb(so) sotoinpcb(so) /* for KAME src sync over BSD's /

	#define INP_SOCKAF(so) so->so_proto->pr_domain->dom_family

	#define INP_CHECK_SOCKAF(so, af) (INP_SOCKAF(so) == af)

	#ifdef _KERNEL
	VNET_DECLARE(int, ipport_reservedhigh);
	VNET_DECLARE(int, ipport_reservedlow);
	VNET_DECLARE(int, ipport_lowfirstauto);
	VNET_DECLARE(int, ipport_lowlastauto);
	VNET_DECLARE(int, ipport_firstauto);
	VNET_DECLARE(int, ipport_lastauto);
	VNET_DECLARE(int, ipport_hifirstauto);
	VNET_DECLARE(int, ipport_hilastauto);
	VNET_DECLARE(int, ipport_randomized);
	VNET_DECLARE(int, ipport_randomcps);
	VNET_DECLARE(int, ipport_randomtime);
	VNET_DECLARE(int, ipport_stoprandom);
	VNET_DECLARE(int, ipport_tcpallocs);

	#define V_ipport_reservedhigh VNET(ipport_reservedhigh)
	#define V_ipport_reservedlow VNET(ipport_reservedlow)
	#define V_ipport_lowfirstauto VNET(ipport_lowfirstauto)
	#define V_ipport_lowlastauto VNET(ipport_lowlastauto)
	#define V_ipport_firstauto VNET(ipport_firstauto)
	#define V_ipport_lastauto VNET(ipport_lastauto)
	#define V_ipport_hifirstauto VNET(ipport_hifirstauto)
	#define V_ipport_hilastauto VNET(ipport_hilastauto)
	#define V_ipport_randomized VNET(ipport_randomized)
	#define V_ipport_randomcps VNET(ipport_randomcps)
	#define V_ipport_randomtime VNET(ipport_randomtime)
	#define V_ipport_stoprandom VNET(ipport_stoprandom)
	#define V_ipport_tcpallocs VNET(ipport_tcpallocs)

	void in_pcbinfo_destroy(struct inpcbinfo *);
	void in_pcbinfo_init(struct inpcbinfo , const char , struct inpcbhead *,
	int, int, char *, uma_init, uma_fini, uint32_t);

	void in_pcbpurgeif0(struct inpcbinfo , struct ifnet );
	int in_pcballoc(struct socket , struct inpcbinfo );
	int in_pcbbind(struct inpcb , struct sockaddr , struct ucred *);
	int in_pcb_lport(struct inpcb , struct in_addr , u_short *,
	struct ucred *, int);
	int in_pcbbind_setup(struct inpcb , struct sockaddr , in_addr_t *,
	u_short , struct ucred );
	int in_pcbconnect(struct inpcb , struct sockaddr , struct ucred *);
	int in_pcbconnect_setup(struct inpcb , struct sockaddr , in_addr_t *,
	u_short , in_addr_t , u_short , struct inpcb *,
	struct ucred *);
	void in_pcbdetach(struct inpcb *);
	void in_pcbdisconnect(struct inpcb *);
	void in_pcbdrop(struct inpcb *);
	void in_pcbfree(struct inpcb *);
	int in_pcbinshash(struct inpcb *);
	struct inpcb *
	in_pcblookup_local(struct inpcbinfo *,
	struct in_addr, u_short, int, struct ucred *);
	struct inpcb *
	in_pcblookup_hash(struct inpcbinfo *, struct in_addr, u_int,
	struct in_addr, u_int, int, struct ifnet *);
	void in_pcbnotifyall(struct inpcbinfo *pcbinfo, struct in_addr,
	int, struct inpcb ()(struct inpcb *, int));
	void in_pcbref(struct inpcb *);
	void in_pcbrehash(struct inpcb *);
	int in_pcbrele(struct inpcb *);
	+int in_pcbrele_rlocked(struct inpcb *);
	+int in_pcbrele_wlocked(struct inpcb *);
	void in_pcbsetsolabel(struct socket *so);
	int in_getpeeraddr(struct socket so, struct sockaddr *nam);
	int in_getsockaddr(struct socket so, struct sockaddr *nam);
	struct sockaddr *
	in_sockaddr(in_port_t port, struct in_addr *addr);
	void in_pcbsosetlabel(struct socket *so);
	#endif /* _KERNEL */

	#endif /* !_NETINET_IN_PCB_H_ */

File Metadata

Mime Type: text/x-diff
Expires: Sun, Sep 13, 5:35 PM (15 h, 45 m)
Storage Engine: blob
Storage Format: Raw Data
Storage Handle: 29575794
Default Alt Text: (78 KB)

No OneTemporaryActions

View Options

File Metadata

Event Timeline

No OneTemporary
Actions