Home | History | Annotate | Line # | Download | only in net
if.h revision 1.280
      1 /*	$NetBSD: if.h,v 1.280 2020/02/01 21:59:39 thorpej Exp $	*/
      2 
      3 /*-
      4  * Copyright (c) 1999, 2000, 2001 The NetBSD Foundation, Inc.
      5  * All rights reserved.
      6  *
      7  * This code is derived from software contributed to The NetBSD Foundation
      8  * by William Studenmund and Jason R. Thorpe.
      9  *
     10  * Redistribution and use in source and binary forms, with or without
     11  * modification, are permitted provided that the following conditions
     12  * are met:
     13  * 1. Redistributions of source code must retain the above copyright
     14  *    notice, this list of conditions and the following disclaimer.
     15  * 2. Redistributions in binary form must reproduce the above copyright
     16  *    notice, this list of conditions and the following disclaimer in the
     17  *    documentation and/or other materials provided with the distribution.
     18  *
     19  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
     20  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
     21  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
     22  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
     23  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
     24  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
     25  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
     26  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
     27  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
     28  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
     29  * POSSIBILITY OF SUCH DAMAGE.
     30  */
     31 
     32 /*
     33  * Copyright (c) 1982, 1986, 1989, 1993
     34  *	The Regents of the University of California.  All rights reserved.
     35  *
     36  * Redistribution and use in source and binary forms, with or without
     37  * modification, are permitted provided that the following conditions
     38  * are met:
     39  * 1. Redistributions of source code must retain the above copyright
     40  *    notice, this list of conditions and the following disclaimer.
     41  * 2. Redistributions in binary form must reproduce the above copyright
     42  *    notice, this list of conditions and the following disclaimer in the
     43  *    documentation and/or other materials provided with the distribution.
     44  * 3. Neither the name of the University nor the names of its contributors
     45  *    may be used to endorse or promote products derived from this software
     46  *    without specific prior written permission.
     47  *
     48  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
     49  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
     50  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
     51  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
     52  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
     53  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
     54  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
     55  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
     56  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
     57  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
     58  * SUCH DAMAGE.
     59  *
     60  *	@(#)if.h	8.3 (Berkeley) 2/9/95
     61  */
     62 
     63 #ifndef _NET_IF_H_
     64 #define _NET_IF_H_
     65 
     66 /*
     67  * Temporary, to allow for a quick disable if problems are discovered
     68  * during a transition period.
     69  */
     70 #define	__IF_STATS_PERCPU
     71 
     72 #if !defined(_KERNEL) && !defined(_STANDALONE)
     73 #include <stdbool.h>
     74 #endif
     75 
     76 #include <sys/featuretest.h>
     77 
     78 /*
     79  * Length of interface external name, including terminating '\0'.
     80  * Note: this is the same size as a generic device's external name.
     81  */
     82 #define IF_NAMESIZE 16
     83 
     84 /*
     85  * Length of interface description, including terminating '\0'.
     86  */
     87 #define	IFDESCRSIZE	64
     88 
     89 #if defined(_NETBSD_SOURCE)
     90 
     91 #include <sys/socket.h>
     92 #include <sys/queue.h>
     93 #include <sys/mutex.h>
     94 
     95 #include <net/dlt.h>
     96 #include <net/pfil.h>
     97 #ifdef _KERNEL
     98 #include <net/pktqueue.h>
     99 #include <sys/pslist.h>
    100 #include <sys/pserialize.h>
    101 #include <sys/psref.h>
    102 #include <sys/module_hook.h>
    103 #endif
    104 
    105 /*
    106  * Always include ALTQ glue here -- we use the ALTQ interface queue
    107  * structure even when ALTQ is not configured into the kernel so that
    108  * the size of struct ifnet does not changed based on the option.  The
    109  * ALTQ queue structure is API-compatible with the legacy ifqueue.
    110  */
    111 #include <altq/if_altq.h>
    112 
    113 /*
    114  * Structures defining a network interface, providing a packet
    115  * transport mechanism (ala level 0 of the PUP protocols).
    116  *
    117  * Each interface accepts output datagrams of a specified maximum
    118  * length, and provides higher level routines with input datagrams
    119  * received from its medium.
    120  *
    121  * Output occurs when the routine if_output is called, with four parameters:
    122  *	(*ifp->if_output)(ifp, m, dst, rt)
    123  * Here m is the mbuf chain to be sent and dst is the destination address.
    124  * The output routine encapsulates the supplied datagram if necessary,
    125  * and then transmits it on its medium.
    126  *
    127  * On input, each interface unwraps the data received by it, and either
    128  * places it on the input queue of a internetwork datagram routine
    129  * and posts the associated software interrupt, or passes the datagram to a raw
    130  * packet input routine.
    131  *
    132  * Routines exist for locating interfaces by their addresses
    133  * or for locating a interface on a certain network, as well as more general
    134  * routing and gateway routines maintaining information used to locate
    135  * interfaces.  These routines live in the files if.c and route.c
    136  */
    137 #include <sys/time.h>
    138 
    139 #if defined(_KERNEL_OPT)
    140 #include "opt_compat_netbsd.h"
    141 #include "opt_gateway.h"
    142 #endif
    143 
    144 struct mbuf;
    145 struct proc;
    146 struct rtentry;
    147 struct socket;
    148 struct ether_header;
    149 struct ifaddr;
    150 struct ifnet;
    151 struct rt_addrinfo;
    152 
    153 #define	IFNAMSIZ	IF_NAMESIZE
    154 
    155 /*
    156  * Structure describing a `cloning' interface.
    157  */
    158 struct if_clone {
    159 	LIST_ENTRY(if_clone) ifc_list;	/* on list of cloners */
    160 	const char *ifc_name;		/* name of device, e.g. `gif' */
    161 	size_t ifc_namelen;		/* length of name */
    162 
    163 	int	(*ifc_create)(struct if_clone *, int);
    164 	int	(*ifc_destroy)(struct ifnet *);
    165 };
    166 
    167 #define	IF_CLONE_INITIALIZER(name, create, destroy)			\
    168 	{ { NULL, NULL }, name, sizeof(name) - 1, create, destroy }
    169 
    170 /*
    171  * Structure used to query names of interface cloners.
    172  */
    173 struct if_clonereq {
    174 	int	ifcr_total;		/* total cloners (out) */
    175 	int	ifcr_count;		/* room for this many in user buffer */
    176 	char	*ifcr_buffer;		/* buffer for cloner names */
    177 };
    178 
    179 /*
    180  * Structure defining statistics and other data kept regarding a network
    181  * interface.
    182  *
    183  * Only used for exporting data from the interface.
    184  */
    185 struct if_data {
    186 	/* generic interface information */
    187 	u_char	ifi_type;		/* ethernet, tokenring, etc. */
    188 	u_char	ifi_addrlen;		/* media address length */
    189 	u_char	ifi_hdrlen;		/* media header length */
    190 	int	ifi_link_state;		/* current link state */
    191 	uint64_t ifi_mtu;		/* maximum transmission unit */
    192 	uint64_t ifi_metric;		/* routing metric (external only) */
    193 	uint64_t ifi_baudrate;		/* linespeed */
    194 	/* volatile statistics */
    195 	uint64_t ifi_ipackets;		/* packets received on interface */
    196 	uint64_t ifi_ierrors;		/* input errors on interface */
    197 	uint64_t ifi_opackets;		/* packets sent on interface */
    198 	uint64_t ifi_oerrors;		/* output errors on interface */
    199 	uint64_t ifi_collisions;	/* collisions on csma interfaces */
    200 	uint64_t ifi_ibytes;		/* total number of octets received */
    201 	uint64_t ifi_obytes;		/* total number of octets sent */
    202 	uint64_t ifi_imcasts;		/* packets received via multicast */
    203 	uint64_t ifi_omcasts;		/* packets sent via multicast */
    204 	uint64_t ifi_iqdrops;		/* dropped on input, this interface */
    205 	uint64_t ifi_noproto;		/* destined for unsupported protocol */
    206 	struct	timespec ifi_lastchange;/* last operational state change */
    207 };
    208 
    209 /*
    210  * Values for if_link_state.
    211  */
    212 #define	LINK_STATE_UNKNOWN	0	/* link invalid/unknown */
    213 #define	LINK_STATE_DOWN		1	/* link is down */
    214 #define	LINK_STATE_UP		2	/* link is up */
    215 
    216 /*
    217  * Structure defining a queue for a network interface.
    218  */
    219 struct ifqueue {
    220 	struct		mbuf *ifq_head;
    221 	struct		mbuf *ifq_tail;
    222 	int		ifq_len;
    223 	int		ifq_maxlen;
    224 	int		ifq_drops;
    225 	kmutex_t	*ifq_lock;
    226 };
    227 
    228 #ifdef _KERNEL
    229 #include <sys/percpu.h>
    230 #include <sys/callout.h>
    231 #include <sys/rwlock.h>
    232 
    233 #endif /* _KERNEL */
    234 
    235 /*
    236  * Structure defining a queue for a network interface.
    237  *
    238  * (Would like to call this struct ``if'', but C isn't PL/1.)
    239  */
    240 TAILQ_HEAD(ifnet_head, ifnet);		/* the actual queue head */
    241 
    242 struct bridge_softc;
    243 struct bridge_iflist;
    244 struct callout;
    245 struct krwlock;
    246 struct if_percpuq;
    247 struct if_deferred_start;
    248 struct in6_multi;
    249 
    250 typedef unsigned short if_index_t;
    251 
    252 /*
    253  * Interface.  Field markings and the corresponding locks:
    254  *
    255  * i:	IFNET_LOCK (a.k.a., if_ioctl_lock)
    256  * q:	ifq_lock (struct ifaltq)
    257  * a:	if_afdata_lock
    258  * 6:	in6_multilock (global lock)
    259  * ::	unlocked, stable
    260  * ?:	unknown, maybe unsafe
    261  *
    262  * Lock order: IFNET_LOCK => in6_multilock => if_afdata_lock => ifq_lock
    263  *   Note that currently if_afdata_lock and ifq_lock aren't held
    264  *   at the same time, but define the order anyway.
    265  *
    266  * Lock order of IFNET_LOCK with other locks:
    267  *     softnet_lock => solock => IFNET_LOCK => ND6_LOCK, in_multilock
    268  */
    269 typedef struct ifnet {
    270 	void		*if_softc;	/* :: lower-level data for this if */
    271 	/* DEPRECATED. Keep it to avoid breaking kvm(3) users */
    272 	TAILQ_ENTRY(ifnet)
    273 			if_list;	/* i: all struct ifnets are chained */
    274 	TAILQ_HEAD(, ifaddr)
    275 			if_addrlist;	/* i: linked list of addresses per if */
    276 	char		if_xname[IFNAMSIZ];
    277 					/* :: external name (name + unit) */
    278 	int		if_pcount;	/* i: number of promiscuous listeners */
    279 	struct bpf_if	*if_bpf;	/* :: packet filter structure */
    280 	if_index_t	if_index;	/* :: numeric abbreviation for this if */
    281 	short		if_timer;	/* ?: time 'til if_slowtimo called */
    282 	unsigned short	if_flags;	/* i: up/down, broadcast, etc. */
    283 	short		if_extflags;	/* :: if_output MP-safe, etc. */
    284 #ifdef __IF_STATS_PERCPU
    285 	u_char		if_type;	/* :: ethernet, tokenring, etc. */
    286 	u_char		if_addrlen;	/* :: media address length */
    287 	u_char		if_hdrlen;	/* :: media header length */
    288 	/* XXX audit :? fields here. */
    289 	int		if_link_state;	/* :? current link state */
    290 	uint64_t	if_mtu;		/* :? maximum transmission unit */
    291 	uint64_t	if_metric;	/* :? routing metric (external only) */
    292 	uint64_t	if_baudrate;	/* :? linespeed */
    293 	struct timespec	if_lastchange;	/* :? last operational state change */
    294 #ifdef _KERNEL
    295 	percpu_t	*if_stats;	/* :: statistics */
    296 #else
    297 	void		*if_stats;	/* opaque to user-space */
    298 #endif /* _KERNEL */
    299 #else /* ! __IF_STATS_PERCPU */
    300 	struct if_data	if_data;	/* ?: statistics and other data */
    301 #endif /* __IF_STATS_PERCPU */
    302 	/*
    303 	 * Procedure handles.  If you add more of these, don't forget the
    304 	 * corresponding NULL stub in if.c.
    305 	 */
    306 	int		(*if_output)	/* :: output routine (enqueue) */
    307 			    (struct ifnet *, struct mbuf *, const struct sockaddr *,
    308 			     const struct rtentry *);
    309 	void		(*_if_input)	/* :: input routine (from h/w driver) */
    310 			    (struct ifnet *, struct mbuf *);
    311 	void		(*if_start)	/* :: initiate output routine */
    312 			    (struct ifnet *);
    313 	int		(*if_transmit)	/* :: output routine, must be MP-safe */
    314 			    (struct ifnet *, struct mbuf *);
    315 	int		(*if_ioctl)	/* :: ioctl routine */
    316 			    (struct ifnet *, u_long, void *);
    317 	int		(*if_init)	/* :: init routine */
    318 			    (struct ifnet *);
    319 	void		(*if_stop)	/* :: stop routine */
    320 			    (struct ifnet *, int);
    321 	void		(*if_slowtimo)	/* :: timer routine */
    322 			    (struct ifnet *);
    323 #define	if_watchdog	if_slowtimo
    324 	void		(*if_drain)	/* :: routine to release resources */
    325 			    (struct ifnet *);
    326 	struct ifaltq	if_snd;		/* q: output queue (includes altq) */
    327 	struct ifaddr	*if_dl;		/* i: identity of this interface. */
    328 	const struct sockaddr_dl
    329 			*if_sadl;	/* i: pointer to sockaddr_dl of if_dl */
    330 	/*
    331 	 * May be NULL.  If not NULL, it is the address assigned
    332 	 * to the interface by the manufacturer, so it very likely
    333 	 * to be unique.  It MUST NOT be deleted.  It is highly
    334 	 * suitable for deriving the EUI64 for the interface.
    335 	 */
    336 	struct ifaddr	*if_hwdl;	/* i: h/w identity */
    337 	const uint8_t	*if_broadcastaddr;
    338 					/* :: linklevel broadcast bytestring */
    339 	struct bridge_softc
    340 			*if_bridge;	/* i: bridge glue */
    341 	struct bridge_iflist
    342 			*if_bridgeif;	/* i: shortcut to interface list entry */
    343 	int		if_dlt;		/* :: data link type (<net/dlt.h>) */
    344 	pfil_head_t *	if_pfil;	/* :: filtering point */
    345 	uint64_t	if_capabilities;
    346 					/* i: interface capabilities */
    347 	uint64_t	if_capenable;	/* i: capabilities enabled */
    348 	union {
    349 		void *		carp_s;	/* carp structure (used by !carp ifs) */
    350 		struct ifnet	*carp_d;/* ptr to carpdev (used by carp ifs) */
    351 	}		if_carp_ptr;	/* ?: */
    352 #define if_carp		if_carp_ptr.carp_s
    353 #define if_carpdev	if_carp_ptr.carp_d
    354 	/*
    355 	 * These are pre-computed based on an interfaces enabled
    356 	 * capabilities, for speed elsewhere.
    357 	 */
    358 	int		if_csum_flags_tx;
    359 					/* i: M_CSUM_* flags for Tx */
    360 	int		if_csum_flags_rx;
    361 					/* i: M_CSUM_* flags for Rx */
    362 
    363 	void		*if_afdata[AF_MAX];
    364 					/* a: */
    365 	struct mowner	*if_mowner;	/* ?: who owns mbufs for this interface */
    366 
    367 	void		*if_agrprivate;	/* ?: used only when #if NAGR > 0 */
    368 	void		*if_npf_private;/* ?: associated NPF context */
    369 
    370 	/*
    371 	 * pf specific data, used only when #if NPF > 0.
    372 	 */
    373 	void		*if_pf_kif;	/* ?: pf interface abstraction */
    374 	void		*if_pf_groups;	/* ?: pf interface groups */
    375 	/*
    376 	 * During an ifnet's lifetime, it has only one if_index, but
    377 	 * and if_index is not sufficient to identify an ifnet
    378 	 * because during the lifetime of the system, many ifnets may occupy a
    379 	 * given if_index.  Let us tell different ifnets at the same
    380 	 * if_index apart by their if_index_gen, a unique number that each ifnet
    381 	 * is assigned when it if_attach()s.  Now, the kernel can use the
    382 	 * pair (if_index, if_index_gen) as a weak reference to an ifnet.
    383 	 */
    384 	uint64_t	if_index_gen;	/* :: generation number for the ifnet
    385 					 * at if_index: if two ifnets' index
    386 					 * and generation number are both the
    387 					 * same, they are the same ifnet.
    388 					 */
    389 	struct sysctllog
    390 			*if_sysctl_log;	/* :: */
    391 	int		(*if_initaddr)  /* :: */
    392 			    (struct ifnet *, struct ifaddr *, bool);
    393 	int		(*if_mcastop)	/* :: */
    394 			    (struct ifnet *, const unsigned long,
    395 			    const struct sockaddr *);
    396 	int		(*if_setflags)	/* :: */
    397 			    (struct ifnet *, const u_short);
    398 	kmutex_t	*if_ioctl_lock;	/* :: */
    399 	char		*if_description;	/* i: interface description */
    400 #ifdef _KERNEL /* XXX kvm(3) */
    401 	struct callout	*if_slowtimo_ch;/* :: */
    402 	struct krwlock	*if_afdata_lock;/* :: */
    403 	struct if_percpuq
    404 			*if_percpuq;	/* :: we should remove it in the future */
    405 	void		*if_link_si;	/* :: softint to handle link state changes */
    406 	uint16_t	if_link_queue;	/* q: masked link state change queue */
    407 	struct pslist_entry
    408 			if_pslist_entry;/* i: */
    409 	struct psref_target
    410 			if_psref;	/* :: */
    411 	struct pslist_head
    412 			if_addr_pslist;	/* i: */
    413 	struct if_deferred_start
    414 			*if_deferred_start;
    415 					/* :: */
    416 	/* XXX should be protocol independent */
    417 	LIST_HEAD(, in6_multi)
    418 			if_multiaddrs;	/* 6: */
    419 #endif
    420 } ifnet_t;
    421 
    422 #include <net/if_stats.h>
    423 
    424 #ifndef __IF_STATS_PERCPU
    425 #define	if_mtu		if_data.ifi_mtu
    426 #define	if_type		if_data.ifi_type
    427 #define	if_addrlen	if_data.ifi_addrlen
    428 #define	if_hdrlen	if_data.ifi_hdrlen
    429 #define	if_metric	if_data.ifi_metric
    430 #define	if_link_state	if_data.ifi_link_state
    431 #define	if_baudrate	if_data.ifi_baudrate
    432 #define	if_ipackets	if_data.ifi_ipackets
    433 #define	if_ierrors	if_data.ifi_ierrors
    434 #define	if_opackets	if_data.ifi_opackets
    435 #define	if_oerrors	if_data.ifi_oerrors
    436 #define	if_collisions	if_data.ifi_collisions
    437 #define	if_ibytes	if_data.ifi_ibytes
    438 #define	if_obytes	if_data.ifi_obytes
    439 #define	if_imcasts	if_data.ifi_imcasts
    440 #define	if_omcasts	if_data.ifi_omcasts
    441 #define	if_iqdrops	if_data.ifi_iqdrops
    442 #define	if_noproto	if_data.ifi_noproto
    443 #define	if_lastchange	if_data.ifi_lastchange
    444 #endif /* __IF_STATS_PERCPU */
    445 #define	if_name(ifp)	((ifp)->if_xname)
    446 
    447 #define	IFF_UP		0x0001		/* interface is up */
    448 #define	IFF_BROADCAST	0x0002		/* broadcast address valid */
    449 #define	IFF_DEBUG	0x0004		/* turn on debugging */
    450 #define	IFF_LOOPBACK	0x0008		/* is a loopback net */
    451 #define	IFF_POINTOPOINT	0x0010		/* interface is point-to-point link */
    452 /*			0x0020		   was IFF_NOTRAILERS */
    453 #define	IFF_RUNNING	0x0040		/* resources allocated */
    454 #define	IFF_NOARP	0x0080		/* no address resolution protocol */
    455 #define	IFF_PROMISC	0x0100		/* receive all packets */
    456 #define	IFF_ALLMULTI	0x0200		/* receive all multicast packets */
    457 #define	IFF_OACTIVE	0x0400		/* transmission in progress */
    458 #define	IFF_SIMPLEX	0x0800		/* can't hear own transmissions */
    459 #define	IFF_LINK0	0x1000		/* per link layer defined bit */
    460 #define	IFF_LINK1	0x2000		/* per link layer defined bit */
    461 #define	IFF_LINK2	0x4000		/* per link layer defined bit */
    462 #define	IFF_MULTICAST	0x8000		/* supports multicast */
    463 
    464 #define	IFEF_MPSAFE			__BIT(0)	/* handlers can run in parallel (see below) */
    465 #define	IFEF_NO_LINK_STATE_CHANGE	__BIT(1)	/* doesn't use link state interrupts */
    466 
    467 /*
    468  * The guidelines for converting an interface to IFEF_MPSAFE are as follows
    469  *
    470  * Enabling IFEF_MPSAFE on an interface suppresses taking KERNEL_LOCK when
    471  * calling the following handlers:
    472  * - if_start
    473  *   - Note that if_transmit is always called without KERNEL_LOCK
    474  * - if_output
    475  * - if_ioctl
    476  * - if_init
    477  * - if_stop
    478  *
    479  * This means that an interface with IFEF_MPSAFE must make the above handlers
    480  * MP-safe or take KERNEL_LOCK by itself inside handlers that aren't MP-safe
    481  * yet.
    482  *
    483  * There are some additional restrictions to access member variables of struct
    484  * ifnet:
    485  * - if_flags
    486  *   - Must be updated with holding IFNET_LOCK
    487  *   - You cannot use the flag in Tx/Rx paths anymore because there is no
    488  *     synchronization on the flag except for IFNET_LOCK
    489  *   - Note that IFNET_LOCK can't be taken in softint because it's known
    490  *     that it causes a deadlock
    491  *     - Some synchronization mechanisms such as pserialize_perform are called
    492  *       with IFNET_LOCK and also require context switches on every CPUs
    493  *       that mean softints finish so trying to take IFNET_LOCK in softint
    494  *       might block on IFNET_LOCK and prevent such synchronization mechanisms
    495  *       from being completed
    496  *     - Currently the deadlock occurs only if NET_MPSAFE is enabled, however,
    497  *       we should deal with the restriction because NET_MPSAFE will be enabled
    498  *       by default in the future
    499  * - if_watchdog and if_timer
    500  *   - The watchdog framework works only for non-IFEF_MPSAFE interfaces
    501  *     that rely on KERNEL_LOCK
    502  *   - Interfaces with IFEF_MPSAFE have to provide its own watchdog mechanism
    503  *     if needed
    504  *     - Keep if_watchdog NULL when calling if_attach
    505  */
    506 
    507 #ifdef _KERNEL
    508 static __inline bool
    509 if_is_mpsafe(struct ifnet *ifp)
    510 {
    511 
    512 	return ((ifp->if_extflags & IFEF_MPSAFE) != 0);
    513 }
    514 
    515 static __inline int
    516 if_output_lock(struct ifnet *cifp, struct ifnet *ifp, struct mbuf *m,
    517     const struct sockaddr *dst, const struct rtentry *rt)
    518 {
    519 
    520 	if (if_is_mpsafe(cifp)) {
    521 		return (*cifp->if_output)(ifp, m, dst, rt);
    522 	} else {
    523 		int ret;
    524 
    525 		KERNEL_LOCK(1, NULL);
    526 		ret = (*cifp->if_output)(ifp, m, dst, rt);
    527 		KERNEL_UNLOCK_ONE(NULL);
    528 		return ret;
    529 	}
    530 }
    531 
    532 static __inline void
    533 if_start_lock(struct ifnet *ifp)
    534 {
    535 
    536 	if (if_is_mpsafe(ifp)) {
    537 		(*ifp->if_start)(ifp);
    538 	} else {
    539 		KERNEL_LOCK(1, NULL);
    540 		(*ifp->if_start)(ifp);
    541 		KERNEL_UNLOCK_ONE(NULL);
    542 	}
    543 }
    544 
    545 static __inline bool
    546 if_is_link_state_changeable(struct ifnet *ifp)
    547 {
    548 
    549 	return ((ifp->if_extflags & IFEF_NO_LINK_STATE_CHANGE) == 0);
    550 }
    551 
    552 #define KERNEL_LOCK_IF_IFP_MPSAFE(ifp)					\
    553 	do { if (if_is_mpsafe(ifp)) { KERNEL_LOCK(1, NULL); } } while (0)
    554 #define KERNEL_UNLOCK_IF_IFP_MPSAFE(ifp)				\
    555 	do { if (if_is_mpsafe(ifp)) { KERNEL_UNLOCK_ONE(NULL); } } while (0)
    556 
    557 #define KERNEL_LOCK_UNLESS_IFP_MPSAFE(ifp)				\
    558 	do { if (!if_is_mpsafe(ifp)) { KERNEL_LOCK(1, NULL); } } while (0)
    559 #define KERNEL_UNLOCK_UNLESS_IFP_MPSAFE(ifp)				\
    560 	do { if (!if_is_mpsafe(ifp)) { KERNEL_UNLOCK_ONE(NULL); } } while (0)
    561 
    562 #ifdef _KERNEL_OPT
    563 #include "opt_net_mpsafe.h"
    564 #endif
    565 
    566 /* XXX explore a better place to define */
    567 #ifdef NET_MPSAFE
    568 
    569 #define KERNEL_LOCK_UNLESS_NET_MPSAFE()		do { } while (0)
    570 #define KERNEL_UNLOCK_UNLESS_NET_MPSAFE()	do { } while (0)
    571 
    572 #define SOFTNET_LOCK_UNLESS_NET_MPSAFE()	do { } while (0)
    573 #define SOFTNET_UNLOCK_UNLESS_NET_MPSAFE()	do { } while (0)
    574 
    575 #define SOFTNET_LOCK_IF_NET_MPSAFE()					\
    576 	do { mutex_enter(softnet_lock); } while (0)
    577 #define SOFTNET_UNLOCK_IF_NET_MPSAFE()					\
    578 	do { mutex_exit(softnet_lock); } while (0)
    579 
    580 #else /* NET_MPSAFE */
    581 
    582 #define KERNEL_LOCK_UNLESS_NET_MPSAFE()					\
    583 	do { KERNEL_LOCK(1, NULL); } while (0)
    584 #define KERNEL_UNLOCK_UNLESS_NET_MPSAFE()				\
    585 	do { KERNEL_UNLOCK_ONE(NULL); } while (0)
    586 
    587 #define SOFTNET_LOCK_UNLESS_NET_MPSAFE()				\
    588 	do { mutex_enter(softnet_lock); } while (0)
    589 #define SOFTNET_UNLOCK_UNLESS_NET_MPSAFE()				\
    590 	do { mutex_exit(softnet_lock); } while (0)
    591 
    592 #define SOFTNET_LOCK_IF_NET_MPSAFE()		do { } while (0)
    593 #define SOFTNET_UNLOCK_IF_NET_MPSAFE()		do { } while (0)
    594 
    595 #endif /* NET_MPSAFE */
    596 
    597 #define SOFTNET_KERNEL_LOCK_UNLESS_NET_MPSAFE()				\
    598 	do {								\
    599 		SOFTNET_LOCK_UNLESS_NET_MPSAFE();			\
    600 		KERNEL_LOCK_UNLESS_NET_MPSAFE();			\
    601 	} while (0)
    602 
    603 #define SOFTNET_KERNEL_UNLOCK_UNLESS_NET_MPSAFE()			\
    604 	do {								\
    605 		KERNEL_UNLOCK_UNLESS_NET_MPSAFE();			\
    606 		SOFTNET_UNLOCK_UNLESS_NET_MPSAFE();			\
    607 	} while (0)
    608 
    609 #endif /* _KERNEL */
    610 
    611 #define	IFFBITS \
    612     "\020\1UP\2BROADCAST\3DEBUG\4LOOPBACK\5POINTOPOINT" \
    613     "\7RUNNING\10NOARP\11PROMISC\12ALLMULTI\13OACTIVE\14SIMPLEX" \
    614     "\15LINK0\16LINK1\17LINK2\20MULTICAST"
    615 
    616 /* flags set internally only: */
    617 #define	IFF_CANTCHANGE \
    618 	(IFF_BROADCAST|IFF_POINTOPOINT|IFF_RUNNING|IFF_OACTIVE|\
    619 	    IFF_SIMPLEX|IFF_MULTICAST|IFF_ALLMULTI|IFF_PROMISC)
    620 
    621 /*
    622  * Some convenience macros used for setting ifi_baudrate.
    623  */
    624 #define	IF_Kbps(x)	((x) * 1000ULL)			/* kilobits/sec. */
    625 #define	IF_Mbps(x)	(IF_Kbps((x) * 1000ULL))	/* megabits/sec. */
    626 #define	IF_Gbps(x)	(IF_Mbps((x) * 1000ULL))	/* gigabits/sec. */
    627 
    628 /* Capabilities that interfaces can advertise. */
    629 					/* 0x01 .. 0x40 were previously used */
    630 #define	IFCAP_TSOv4		0x00080	/* can do TCPv4 segmentation offload */
    631 #define	IFCAP_CSUM_IPv4_Rx	0x00100	/* can do IPv4 header checksums (Rx) */
    632 #define	IFCAP_CSUM_IPv4_Tx	0x00200	/* can do IPv4 header checksums (Tx) */
    633 #define	IFCAP_CSUM_TCPv4_Rx	0x00400	/* can do IPv4/TCP checksums (Rx) */
    634 #define	IFCAP_CSUM_TCPv4_Tx	0x00800	/* can do IPv4/TCP checksums (Tx) */
    635 #define	IFCAP_CSUM_UDPv4_Rx	0x01000	/* can do IPv4/UDP checksums (Rx) */
    636 #define	IFCAP_CSUM_UDPv4_Tx	0x02000	/* can do IPv4/UDP checksums (Tx) */
    637 #define	IFCAP_CSUM_TCPv6_Rx	0x04000	/* can do IPv6/TCP checksums (Rx) */
    638 #define	IFCAP_CSUM_TCPv6_Tx	0x08000	/* can do IPv6/TCP checksums (Tx) */
    639 #define	IFCAP_CSUM_UDPv6_Rx	0x10000	/* can do IPv6/UDP checksums (Rx) */
    640 #define	IFCAP_CSUM_UDPv6_Tx	0x20000	/* can do IPv6/UDP checksums (Tx) */
    641 #define	IFCAP_TSOv6		0x40000	/* can do TCPv6 segmentation offload */
    642 #define	IFCAP_LRO		0x80000	/* can do Large Receive Offload */
    643 #define	IFCAP_MASK		0xfff80 /* currently valid capabilities */
    644 
    645 #define	IFCAPBITS		\
    646 	"\020"			\
    647 	"\10TSO4"		\
    648 	"\11IP4CSUM_Rx"		\
    649 	"\12IP4CSUM_Tx"		\
    650 	"\13TCP4CSUM_Rx"	\
    651 	"\14TCP4CSUM_Tx"	\
    652 	"\15UDP4CSUM_Rx"	\
    653 	"\16UDP4CSUM_Tx"	\
    654 	"\17TCP6CSUM_Rx"	\
    655 	"\20TCP6CSUM_Tx"	\
    656 	"\21UDP6CSUM_Rx"	\
    657 	"\22UDP6CSUM_Tx"	\
    658 	"\23TSO6"		\
    659 	"\24LRO"		\
    660 
    661 #define	IF_AFDATA_LOCK_INIT(ifp)	\
    662 	do {(ifp)->if_afdata_lock = rw_obj_alloc();} while (0)
    663 
    664 #define	IF_AFDATA_LOCK_DESTROY(ifp)	rw_obj_free((ifp)->if_afdata_lock)
    665 
    666 #define	IF_AFDATA_WLOCK(ifp)	rw_enter((ifp)->if_afdata_lock, RW_WRITER)
    667 #define	IF_AFDATA_RLOCK(ifp)	rw_enter((ifp)->if_afdata_lock, RW_READER)
    668 #define	IF_AFDATA_WUNLOCK(ifp)	rw_exit((ifp)->if_afdata_lock)
    669 #define	IF_AFDATA_RUNLOCK(ifp)	rw_exit((ifp)->if_afdata_lock)
    670 #define	IF_AFDATA_LOCK(ifp)	IF_AFDATA_WLOCK(ifp)
    671 #define	IF_AFDATA_UNLOCK(ifp)	IF_AFDATA_WUNLOCK(ifp)
    672 #define	IF_AFDATA_TRYLOCK(ifp)	rw_tryenter((ifp)->if_afdata_lock, RW_WRITER)
    673 
    674 #define	IF_AFDATA_LOCK_ASSERT(ifp)	\
    675 	KASSERT(rw_lock_held((ifp)->if_afdata_lock))
    676 #define	IF_AFDATA_RLOCK_ASSERT(ifp)	\
    677 	KASSERT(rw_read_held((ifp)->if_afdata_lock))
    678 #define	IF_AFDATA_WLOCK_ASSERT(ifp)	\
    679 	KASSERT(rw_write_held((ifp)->if_afdata_lock))
    680 
    681 /*
    682  * Output queues (ifp->if_snd) and internetwork datagram level (pup level 1)
    683  * input routines have queues of messages stored on ifqueue structures
    684  * (defined above).  Entries are added to and deleted from these structures
    685  * by these macros, which should be called with ipl raised to splnet().
    686  */
    687 #define	IF_QFULL(ifq)		((ifq)->ifq_len >= (ifq)->ifq_maxlen)
    688 #define	IF_DROP(ifq)		((ifq)->ifq_drops++)
    689 #define	IF_ENQUEUE(ifq, m) do { \
    690 	(m)->m_nextpkt = 0; \
    691 	if ((ifq)->ifq_tail == 0) \
    692 		(ifq)->ifq_head = m; \
    693 	else \
    694 		(ifq)->ifq_tail->m_nextpkt = m; \
    695 	(ifq)->ifq_tail = m; \
    696 	(ifq)->ifq_len++; \
    697 } while (/*CONSTCOND*/0)
    698 #define	IF_PREPEND(ifq, m) do { \
    699 	(m)->m_nextpkt = (ifq)->ifq_head; \
    700 	if ((ifq)->ifq_tail == 0) \
    701 		(ifq)->ifq_tail = (m); \
    702 	(ifq)->ifq_head = (m); \
    703 	(ifq)->ifq_len++; \
    704 } while (/*CONSTCOND*/0)
    705 #define	IF_DEQUEUE(ifq, m) do { \
    706 	(m) = (ifq)->ifq_head; \
    707 	if (m) { \
    708 		if (((ifq)->ifq_head = (m)->m_nextpkt) == 0) \
    709 			(ifq)->ifq_tail = 0; \
    710 		(m)->m_nextpkt = 0; \
    711 		(ifq)->ifq_len--; \
    712 	} \
    713 } while (/*CONSTCOND*/0)
    714 #define	IF_POLL(ifq, m)		((m) = (ifq)->ifq_head)
    715 #define	IF_PURGE(ifq)							\
    716 do {									\
    717 	struct mbuf *__m0;						\
    718 									\
    719 	for (;;) {							\
    720 		IF_DEQUEUE((ifq), __m0);				\
    721 		if (__m0 == NULL)					\
    722 			break;						\
    723 		else							\
    724 			m_freem(__m0);					\
    725 	}								\
    726 } while (/*CONSTCOND*/ 0)
    727 #define	IF_IS_EMPTY(ifq)	((ifq)->ifq_len == 0)
    728 
    729 #ifndef IFQ_MAXLEN
    730 #define	IFQ_MAXLEN	256
    731 #endif
    732 #define	IFNET_SLOWHZ	1		/* granularity is 1 second */
    733 
    734 /*
    735  * Structure defining statistics and other data kept regarding an address
    736  * on a network interface.
    737  */
    738 struct ifaddr_data {
    739 	int64_t	ifad_inbytes;
    740 	int64_t	ifad_outbytes;
    741 };
    742 
    743 /*
    744  * The ifaddr structure contains information about one address
    745  * of an interface.  They are maintained by the different address families,
    746  * are allocated and attached when an address is set, and are linked
    747  * together so all addresses for an interface can be located.
    748  */
    749 struct ifaddr {
    750 	struct	sockaddr *ifa_addr;	/* address of interface */
    751 	struct	sockaddr *ifa_dstaddr;	/* other end of p-to-p link */
    752 #define	ifa_broadaddr	ifa_dstaddr	/* broadcast address interface */
    753 	struct	sockaddr *ifa_netmask;	/* used to determine subnet */
    754 	struct	ifnet *ifa_ifp;		/* back-pointer to interface */
    755 	TAILQ_ENTRY(ifaddr) ifa_list;	/* list of addresses for interface */
    756 	struct	ifaddr_data	ifa_data;	/* statistics on the address */
    757 	void	(*ifa_rtrequest)	/* check or clean routes (+ or -)'d */
    758 		        (int, struct rtentry *, const struct rt_addrinfo *);
    759 	u_int	ifa_flags;		/* mostly rt_flags for cloning */
    760 	int	ifa_refcnt;		/* count of references */
    761 	int	ifa_metric;		/* cost of going out this interface */
    762 	struct ifaddr	*(*ifa_getifa)(struct ifaddr *,
    763 			               const struct sockaddr *);
    764 	uint32_t	*ifa_seqno;
    765 	int16_t	ifa_preference;	/* preference level for this address */
    766 #ifdef _KERNEL
    767 	struct pslist_entry     ifa_pslist_entry;
    768 	struct psref_target	ifa_psref;
    769 #endif
    770 };
    771 #define	IFA_ROUTE	RTF_UP	/* (0x01) route installed */
    772 #define	IFA_DESTROYING	0x2
    773 
    774 /*
    775  * Message format for use in obtaining information about interfaces from
    776  * sysctl and the routing socket.  We need to force 64-bit alignment if we
    777  * aren't using compatiblity definitons.
    778  */
    779 #if !defined(_KERNEL) || !defined(COMPAT_RTSOCK)
    780 #define	__align64	__aligned(sizeof(uint64_t))
    781 #else
    782 #define	__align64
    783 #endif
    784 struct if_msghdr {
    785 	u_short	ifm_msglen __align64;
    786 				/* to skip over non-understood messages */
    787 	u_char	ifm_version;	/* future binary compatibility */
    788 	u_char	ifm_type;	/* message type */
    789 	int	ifm_addrs;	/* like rtm_addrs */
    790 	int	ifm_flags;	/* value of if_flags */
    791 	u_short	ifm_index;	/* index for associated ifp */
    792 	struct	if_data ifm_data __align64;
    793 				/* statistics and other data about if */
    794 };
    795 
    796 /*
    797  * Message format for use in obtaining information about interface addresses
    798  * from sysctl and the routing socket.
    799  */
    800 struct ifa_msghdr {
    801 	u_short	ifam_msglen __align64;
    802 				/* to skip over non-understood messages */
    803 	u_char	ifam_version;	/* future binary compatibility */
    804 	u_char	ifam_type;	/* message type */
    805 	u_short	ifam_index;	/* index for associated ifp */
    806 	int	ifam_flags;	/* value of ifa_flags */
    807 	int	ifam_addrs;	/* like rtm_addrs */
    808 	pid_t	ifam_pid;	/* identify sender */
    809 	int	ifam_addrflags;	/* family specific address flags */
    810 	int	ifam_metric;	/* value of ifa_metric */
    811 };
    812 
    813 /*
    814  * Message format announcing the arrival or departure of a network interface.
    815  */
    816 struct if_announcemsghdr {
    817 	u_short	ifan_msglen __align64;
    818 				/* to skip over non-understood messages */
    819 	u_char	ifan_version;	/* future binary compatibility */
    820 	u_char	ifan_type;	/* message type */
    821 	u_short	ifan_index;	/* index for associated ifp */
    822 	char	ifan_name[IFNAMSIZ]; /* if name, e.g. "en0" */
    823 	u_short	ifan_what;	/* what type of announcement */
    824 };
    825 
    826 #define	IFAN_ARRIVAL	0	/* interface arrival */
    827 #define	IFAN_DEPARTURE	1	/* interface departure */
    828 
    829 #undef __align64
    830 
    831 /*
    832  * Interface request structure used for socket
    833  * ioctl's.  All interface ioctl's must have parameter
    834  * definitions which begin with ifr_name.  The
    835  * remainder may be interface specific.
    836  */
    837 struct	ifreq {
    838 	char	ifr_name[IFNAMSIZ];		/* if name, e.g. "en0" */
    839 	union {
    840 		struct	sockaddr ifru_addr;
    841 		struct	sockaddr ifru_dstaddr;
    842 		struct	sockaddr ifru_broadaddr;
    843 		struct	sockaddr_storage ifru_space;
    844 		short	ifru_flags;
    845 		int	ifru_addrflags;
    846 		int	ifru_metric;
    847 		int	ifru_mtu;
    848 		int	ifru_dlt;
    849 		u_int	ifru_value;
    850 		void *	ifru_data;
    851 		struct {
    852 			uint32_t	b_buflen;
    853 			void		*b_buf;
    854 		} ifru_b;
    855 	} ifr_ifru;
    856 #define	ifr_addr	ifr_ifru.ifru_addr	/* address */
    857 #define	ifr_dstaddr	ifr_ifru.ifru_dstaddr	/* other end of p-to-p link */
    858 #define	ifr_broadaddr	ifr_ifru.ifru_broadaddr	/* broadcast address */
    859 #define	ifr_space	ifr_ifru.ifru_space	/* sockaddr_storage */
    860 #define	ifr_flags	ifr_ifru.ifru_flags	/* flags */
    861 #define	ifr_addrflags	ifr_ifru.ifru_addrflags	/* addr flags */
    862 #define	ifr_metric	ifr_ifru.ifru_metric	/* metric */
    863 #define	ifr_mtu		ifr_ifru.ifru_mtu	/* mtu */
    864 #define	ifr_dlt		ifr_ifru.ifru_dlt	/* data link type (DLT_*) */
    865 #define	ifr_value	ifr_ifru.ifru_value	/* generic value */
    866 #define	ifr_media	ifr_ifru.ifru_metric	/* media options (overload) */
    867 #define	ifr_data	ifr_ifru.ifru_data	/* for use by interface
    868 						 * XXX deprecated
    869 						 */
    870 #define	ifr_buf		ifr_ifru.ifru_b.b_buf	/* new interface ioctls */
    871 #define	ifr_buflen	ifr_ifru.ifru_b.b_buflen
    872 #define	ifr_index	ifr_ifru.ifru_value	/* interface index, BSD */
    873 #define	ifr_ifindex	ifr_index		/* interface index, linux */
    874 };
    875 
    876 #ifdef _KERNEL
    877 #define	ifreq_setdstaddr	ifreq_setaddr
    878 #define	ifreq_setbroadaddr	ifreq_setaddr
    879 #define	ifreq_getdstaddr	ifreq_getaddr
    880 #define	ifreq_getbroadaddr	ifreq_getaddr
    881 
    882 static __inline const struct sockaddr *
    883 /*ARGSUSED*/
    884 ifreq_getaddr(u_long cmd, const struct ifreq *ifr)
    885 {
    886 	return &ifr->ifr_addr;
    887 }
    888 #endif /* _KERNEL */
    889 
    890 struct ifcapreq {
    891 	char		ifcr_name[IFNAMSIZ];	/* if name, e.g. "en0" */
    892 	uint64_t	ifcr_capabilities;	/* supported capabiliites */
    893 	uint64_t	ifcr_capenable;		/* capabilities enabled */
    894 };
    895 
    896 struct ifaliasreq {
    897 	char	ifra_name[IFNAMSIZ];		/* if name, e.g. "en0" */
    898 	struct	sockaddr ifra_addr;
    899 	struct	sockaddr ifra_dstaddr;
    900 #define	ifra_broadaddr	ifra_dstaddr
    901 	struct	sockaddr ifra_mask;
    902 };
    903 
    904 struct ifdatareq {
    905 	char	ifdr_name[IFNAMSIZ];		/* if name, e.g. "en0" */
    906 	struct	if_data ifdr_data;
    907 };
    908 
    909 struct ifmediareq {
    910 	char	ifm_name[IFNAMSIZ];	/* if name, e.g. "en0" */
    911 	int	ifm_current;		/* IFMWD: current media options */
    912 	int	ifm_mask;		/* IFMWD: don't care mask */
    913 	int	ifm_status;		/* media status */
    914 	int	ifm_active;		/* IFMWD: active options */
    915 	int	ifm_count;		/* # entries in ifm_ulist
    916 					   array */
    917 	int	*ifm_ulist;		/* array of ifmedia word */
    918 };
    919 
    920 
    921 struct  ifdrv {
    922 	char		ifd_name[IFNAMSIZ];	/* if name, e.g. "en0" */
    923 	unsigned long	ifd_cmd;
    924 	size_t		ifd_len;
    925 	void		*ifd_data;
    926 };
    927 #define IFLINKSTR_QUERYLEN	0x01
    928 #define IFLINKSTR_UNSET		0x02
    929 
    930 /*
    931  * Structure used in SIOCGIFCONF request.
    932  * Used to retrieve interface configuration
    933  * for machine (useful for programs which
    934  * must know all networks accessible).
    935  */
    936 struct	ifconf {
    937 	int	ifc_len;		/* size of associated buffer */
    938 	union {
    939 		void *	ifcu_buf;
    940 		struct	ifreq *ifcu_req;
    941 	} ifc_ifcu;
    942 #define	ifc_buf	ifc_ifcu.ifcu_buf	/* buffer address */
    943 #define	ifc_req	ifc_ifcu.ifcu_req	/* array of structures returned */
    944 };
    945 
    946 /*
    947  * Structure for SIOC[AGD]LIFADDR
    948  */
    949 struct if_laddrreq {
    950 	char iflr_name[IFNAMSIZ];
    951 	unsigned int flags;
    952 #define IFLR_PREFIX	0x8000	/* in: prefix given  out: kernel fills id */
    953 #define IFLR_ACTIVE	0x4000	/* in/out: link-layer address activation */
    954 #define IFLR_FACTORY	0x2000	/* in/out: factory link-layer address */
    955 	unsigned int prefixlen;		/* in/out */
    956 	struct sockaddr_storage addr;	/* in/out */
    957 	struct sockaddr_storage dstaddr; /* out */
    958 };
    959 
    960 /*
    961  * Structure for SIOC[SG]IFADDRPREF
    962  */
    963 struct if_addrprefreq {
    964 	char			ifap_name[IFNAMSIZ];
    965 	int16_t			ifap_preference;	/* in/out */
    966 	struct sockaddr_storage	ifap_addr;		/* in/out */
    967 };
    968 
    969 #include <net/if_arp.h>
    970 
    971 #endif /* _NETBSD_SOURCE */
    972 
    973 #ifdef _KERNEL
    974 #ifdef ALTQ
    975 #define IFQ_ENQUEUE(ifq, m, err)					\
    976 do {									\
    977 	mutex_enter((ifq)->ifq_lock);					\
    978 	if (ALTQ_IS_ENABLED(ifq))					\
    979 		ALTQ_ENQUEUE((ifq), (m), (err));			\
    980 	else {								\
    981 		if (IF_QFULL(ifq)) {					\
    982 			m_freem(m);					\
    983 			(err) = ENOBUFS;				\
    984 		} else {						\
    985 			IF_ENQUEUE((ifq), (m));				\
    986 			(err) = 0;					\
    987 		}							\
    988 	}								\
    989 	if ((err))							\
    990 		(ifq)->ifq_drops++;					\
    991 	mutex_exit((ifq)->ifq_lock);					\
    992 } while (/*CONSTCOND*/ 0)
    993 
    994 #define IFQ_DEQUEUE(ifq, m)						\
    995 do {									\
    996 	mutex_enter((ifq)->ifq_lock);					\
    997 	if (TBR_IS_ENABLED(ifq))					\
    998 		(m) = tbr_dequeue((ifq), ALTDQ_REMOVE);			\
    999 	else if (ALTQ_IS_ENABLED(ifq))					\
   1000 		ALTQ_DEQUEUE((ifq), (m));				\
   1001 	else								\
   1002 		IF_DEQUEUE((ifq), (m));					\
   1003 	mutex_exit((ifq)->ifq_lock);					\
   1004 } while (/*CONSTCOND*/ 0)
   1005 
   1006 #define	IFQ_POLL(ifq, m)						\
   1007 do {									\
   1008 	mutex_enter((ifq)->ifq_lock);					\
   1009 	if (TBR_IS_ENABLED(ifq))					\
   1010 		(m) = tbr_dequeue((ifq), ALTDQ_POLL);			\
   1011 	else if (ALTQ_IS_ENABLED(ifq))					\
   1012 		ALTQ_POLL((ifq), (m));					\
   1013 	else								\
   1014 		IF_POLL((ifq), (m));					\
   1015 	mutex_exit((ifq)->ifq_lock);					\
   1016 } while (/*CONSTCOND*/ 0)
   1017 
   1018 #define	IFQ_PURGE(ifq)							\
   1019 do {									\
   1020 	mutex_enter((ifq)->ifq_lock);					\
   1021 	if (ALTQ_IS_ENABLED(ifq))					\
   1022 		ALTQ_PURGE(ifq);					\
   1023 	else								\
   1024 		IF_PURGE(ifq);						\
   1025 	mutex_exit((ifq)->ifq_lock);					\
   1026 } while (/*CONSTCOND*/ 0)
   1027 
   1028 #define	IFQ_SET_READY(ifq)						\
   1029 do {									\
   1030 	(ifq)->altq_flags |= ALTQF_READY;				\
   1031 } while (/*CONSTCOND*/ 0)
   1032 
   1033 #define	IFQ_CLASSIFY(ifq, m, af)					\
   1034 do {									\
   1035 	KASSERT(((m)->m_flags & M_PKTHDR) != 0);			\
   1036 	mutex_enter((ifq)->ifq_lock);					\
   1037 	if (ALTQ_IS_ENABLED(ifq)) {					\
   1038 		if (ALTQ_NEEDS_CLASSIFY(ifq))				\
   1039 			(m)->m_pkthdr.pattr_class = (*(ifq)->altq_classify) \
   1040 				((ifq)->altq_clfier, (m), (af));	\
   1041 		(m)->m_pkthdr.pattr_af = (af);				\
   1042 		(m)->m_pkthdr.pattr_hdr = mtod((m), void *);		\
   1043 	}								\
   1044 	mutex_exit((ifq)->ifq_lock);					\
   1045 } while (/*CONSTCOND*/ 0)
   1046 #else /* ! ALTQ */
   1047 #define	IFQ_ENQUEUE(ifq, m, err)					\
   1048 do {									\
   1049 	mutex_enter((ifq)->ifq_lock);					\
   1050 	if (IF_QFULL(ifq)) {						\
   1051 		m_freem(m);						\
   1052 		(err) = ENOBUFS;					\
   1053 	} else {							\
   1054 		IF_ENQUEUE((ifq), (m));					\
   1055 		(err) = 0;						\
   1056 	}								\
   1057 	if (err)							\
   1058 		(ifq)->ifq_drops++;					\
   1059 	mutex_exit((ifq)->ifq_lock);					\
   1060 } while (/*CONSTCOND*/ 0)
   1061 
   1062 #define	IFQ_DEQUEUE(ifq, m)						\
   1063 do {									\
   1064 	mutex_enter((ifq)->ifq_lock);					\
   1065 	IF_DEQUEUE((ifq), (m));						\
   1066 	mutex_exit((ifq)->ifq_lock);					\
   1067 } while (/*CONSTCOND*/ 0)
   1068 
   1069 #define	IFQ_POLL(ifq, m)						\
   1070 do {									\
   1071 	mutex_enter((ifq)->ifq_lock);					\
   1072 	IF_POLL((ifq), (m));						\
   1073 	mutex_exit((ifq)->ifq_lock);					\
   1074 } while (/*CONSTCOND*/ 0)
   1075 
   1076 #define	IFQ_PURGE(ifq)							\
   1077 do {									\
   1078 	mutex_enter((ifq)->ifq_lock);					\
   1079 	IF_PURGE(ifq);							\
   1080 	mutex_exit((ifq)->ifq_lock);					\
   1081 } while (/*CONSTCOND*/ 0)
   1082 
   1083 #define	IFQ_SET_READY(ifq)	/* nothing */
   1084 
   1085 #define	IFQ_CLASSIFY(ifq, m, af) /* nothing */
   1086 
   1087 #endif /* ALTQ */
   1088 
   1089 #define IFQ_LOCK_INIT(ifq)	(ifq)->ifq_lock =			\
   1090 	    mutex_obj_alloc(MUTEX_DEFAULT, IPL_NET)
   1091 #define IFQ_LOCK_DESTROY(ifq)	mutex_obj_free((ifq)->ifq_lock)
   1092 #define IFQ_LOCK(ifq)		mutex_enter((ifq)->ifq_lock)
   1093 #define IFQ_UNLOCK(ifq)		mutex_exit((ifq)->ifq_lock)
   1094 
   1095 #define	IFQ_IS_EMPTY(ifq)		IF_IS_EMPTY(ifq)
   1096 #define	IFQ_INC_LEN(ifq)		((ifq)->ifq_len++)
   1097 #define	IFQ_DEC_LEN(ifq)		(--(ifq)->ifq_len)
   1098 #define	IFQ_INC_DROPS(ifq)		((ifq)->ifq_drops++)
   1099 #define	IFQ_SET_MAXLEN(ifq, len)	((ifq)->ifq_maxlen = (len))
   1100 
   1101 #include <sys/mallocvar.h>
   1102 MALLOC_DECLARE(M_IFADDR);
   1103 MALLOC_DECLARE(M_IFMADDR);
   1104 
   1105 int ifreq_setaddr(u_long, struct ifreq *, const struct sockaddr *);
   1106 
   1107 struct ifnet *if_alloc(u_char);
   1108 void if_free(struct ifnet *);
   1109 void if_initname(struct ifnet *, const char *, int);
   1110 struct ifaddr *if_dl_create(const struct ifnet *, const struct sockaddr_dl **);
   1111 void if_activate_sadl(struct ifnet *, struct ifaddr *,
   1112     const struct sockaddr_dl *);
   1113 void	if_set_sadl(struct ifnet *, const void *, u_char, bool);
   1114 void	if_alloc_sadl(struct ifnet *);
   1115 void	if_free_sadl(struct ifnet *, int);
   1116 int	if_initialize(struct ifnet *);
   1117 void	if_register(struct ifnet *);
   1118 int	if_attach(struct ifnet *); /* Deprecated. Use if_initialize and if_register */
   1119 void	if_attachdomain(void);
   1120 void	if_deactivate(struct ifnet *);
   1121 bool	if_is_deactivated(const struct ifnet *);
   1122 void	if_export_if_data(struct ifnet *, struct if_data *, bool);
   1123 void	if_purgeaddrs(struct ifnet *, int, void (*)(struct ifaddr *));
   1124 void	if_detach(struct ifnet *);
   1125 void	if_down(struct ifnet *);
   1126 void	if_down_locked(struct ifnet *);
   1127 void	if_link_state_change(struct ifnet *, int);
   1128 void	if_link_state_change_softint(struct ifnet *, int);
   1129 void	if_up(struct ifnet *);
   1130 void	ifinit(void);
   1131 void	ifinit1(void);
   1132 void	ifinit_post(void);
   1133 int	ifaddrpref_ioctl(struct socket *, u_long, void *, struct ifnet *);
   1134 extern int (*ifioctl)(struct socket *, u_long, void *, struct lwp *);
   1135 int	ifioctl_common(struct ifnet *, u_long, void *);
   1136 int	ifpromisc(struct ifnet *, int);
   1137 int	ifpromisc_locked(struct ifnet *, int);
   1138 int	if_addr_init(ifnet_t *, struct ifaddr *, bool);
   1139 int	if_do_dad(struct ifnet *);
   1140 int	if_mcast_op(ifnet_t *, const unsigned long, const struct sockaddr *);
   1141 int	if_flags_set(struct ifnet *, const u_short);
   1142 int	if_clone_list(int, char *, int *);
   1143 
   1144 struct	ifnet *ifunit(const char *);
   1145 struct	ifnet *if_get(const char *, struct psref *);
   1146 ifnet_t *if_byindex(u_int);
   1147 ifnet_t *_if_byindex(u_int);
   1148 ifnet_t *if_get_byindex(u_int, struct psref *);
   1149 ifnet_t *if_get_bylla(const void *, unsigned char, struct psref *);
   1150 void	if_put(const struct ifnet *, struct psref *);
   1151 void	if_acquire(struct ifnet *, struct psref *);
   1152 #define	if_release	if_put
   1153 
   1154 int if_tunnel_check_nesting(struct ifnet *, struct mbuf *, int);
   1155 percpu_t *if_tunnel_alloc_ro_percpu(void);
   1156 void if_tunnel_free_ro_percpu(percpu_t *);
   1157 void if_tunnel_ro_percpu_rtcache_free(percpu_t *);
   1158 
   1159 struct tunnel_ro {
   1160 	struct route *tr_ro;
   1161 	kmutex_t *tr_lock;
   1162 };
   1163 
   1164 static inline void
   1165 if_tunnel_get_ro(percpu_t *ro_percpu, struct route **ro, kmutex_t **lock)
   1166 {
   1167 	struct tunnel_ro *tro;
   1168 
   1169 	tro = percpu_getref(ro_percpu);
   1170 	*ro = tro->tr_ro;
   1171 	*lock = tro->tr_lock;
   1172 	mutex_enter(*lock);
   1173 }
   1174 
   1175 static inline void
   1176 if_tunnel_put_ro(percpu_t *ro_percpu, kmutex_t *lock)
   1177 {
   1178 
   1179 	mutex_exit(lock);
   1180 	percpu_putref(ro_percpu);
   1181 }
   1182 
   1183 static __inline if_index_t
   1184 if_get_index(const struct ifnet *ifp)
   1185 {
   1186 
   1187 	return ifp != NULL ? ifp->if_index : 0;
   1188 }
   1189 
   1190 bool	if_held(struct ifnet *);
   1191 
   1192 void	if_input(struct ifnet *, struct mbuf *);
   1193 
   1194 struct if_percpuq *
   1195 	if_percpuq_create(struct ifnet *);
   1196 void	if_percpuq_destroy(struct if_percpuq *);
   1197 void
   1198 	if_percpuq_enqueue(struct if_percpuq *, struct mbuf *);
   1199 
   1200 void	if_deferred_start_init(struct ifnet *, void (*)(struct ifnet *));
   1201 void	if_schedule_deferred_start(struct ifnet *);
   1202 
   1203 void ifa_insert(struct ifnet *, struct ifaddr *);
   1204 void ifa_remove(struct ifnet *, struct ifaddr *);
   1205 
   1206 void	ifa_psref_init(struct ifaddr *);
   1207 void	ifa_acquire(struct ifaddr *, struct psref *);
   1208 void	ifa_release(struct ifaddr *, struct psref *);
   1209 bool	ifa_held(struct ifaddr *);
   1210 bool	ifa_is_destroying(struct ifaddr *);
   1211 
   1212 void	ifaref(struct ifaddr *);
   1213 void	ifafree(struct ifaddr *);
   1214 
   1215 struct	ifaddr *ifa_ifwithaddr(const struct sockaddr *);
   1216 struct	ifaddr *ifa_ifwithaddr_psref(const struct sockaddr *, struct psref *);
   1217 struct	ifaddr *ifa_ifwithaf(int);
   1218 struct	ifaddr *ifa_ifwithdstaddr(const struct sockaddr *);
   1219 struct	ifaddr *ifa_ifwithdstaddr_psref(const struct sockaddr *,
   1220 	    struct psref *);
   1221 struct	ifaddr *ifa_ifwithnet(const struct sockaddr *);
   1222 struct	ifaddr *ifa_ifwithnet_psref(const struct sockaddr *, struct psref *);
   1223 struct	ifaddr *ifa_ifwithladdr(const struct sockaddr *);
   1224 struct	ifaddr *ifa_ifwithladdr_psref(const struct sockaddr *, struct psref *);
   1225 struct	ifaddr *ifaof_ifpforaddr(const struct sockaddr *, struct ifnet *);
   1226 struct	ifaddr *ifaof_ifpforaddr_psref(const struct sockaddr *, struct ifnet *,
   1227 	    struct psref *);
   1228 void	link_rtrequest(int, struct rtentry *, const struct rt_addrinfo *);
   1229 void	p2p_rtrequest(int, struct rtentry *, const struct rt_addrinfo *);
   1230 
   1231 void	if_clone_attach(struct if_clone *);
   1232 void	if_clone_detach(struct if_clone *);
   1233 
   1234 int	if_transmit_lock(struct ifnet *, struct mbuf *);
   1235 
   1236 int	ifq_enqueue(struct ifnet *, struct mbuf *);
   1237 int	ifq_enqueue2(struct ifnet *, struct ifqueue *, struct mbuf *);
   1238 
   1239 int	loioctl(struct ifnet *, u_long, void *);
   1240 void	loopattach(int);
   1241 void	loopinit(void);
   1242 int	looutput(struct ifnet *,
   1243 	   struct mbuf *, const struct sockaddr *, const struct rtentry *);
   1244 
   1245 /*
   1246  * These are exported because they're an easy way to tell if
   1247  * an interface is going away without having to burn a flag.
   1248  */
   1249 int	if_nulloutput(struct ifnet *, struct mbuf *,
   1250 	    const struct sockaddr *, const struct rtentry *);
   1251 void	if_nullinput(struct ifnet *, struct mbuf *);
   1252 void	if_nullstart(struct ifnet *);
   1253 int	if_nulltransmit(struct ifnet *, struct mbuf *);
   1254 int	if_nullioctl(struct ifnet *, u_long, void *);
   1255 int	if_nullinit(struct ifnet *);
   1256 void	if_nullstop(struct ifnet *, int);
   1257 void	if_nullslowtimo(struct ifnet *);
   1258 #define	if_nullwatchdog	if_nullslowtimo
   1259 void	if_nulldrain(struct ifnet *);
   1260 #else
   1261 struct if_nameindex {
   1262 	unsigned int	if_index;	/* 1, 2, ... */
   1263 	char		*if_name;	/* null terminated name: "le0", ... */
   1264 };
   1265 
   1266 #include <sys/cdefs.h>
   1267 __BEGIN_DECLS
   1268 unsigned int if_nametoindex(const char *);
   1269 char *	if_indextoname(unsigned int, char *);
   1270 struct	if_nameindex * if_nameindex(void);
   1271 void	if_freenameindex(struct if_nameindex *);
   1272 __END_DECLS
   1273 #endif /* _KERNEL */ /* XXX really ALTQ? */
   1274 
   1275 #ifdef _KERNEL
   1276 
   1277 #define	IFADDR_FIRST(__ifp)		TAILQ_FIRST(&(__ifp)->if_addrlist)
   1278 #define	IFADDR_NEXT(__ifa)		TAILQ_NEXT((__ifa), ifa_list)
   1279 #define	IFADDR_FOREACH(__ifa, __ifp)	TAILQ_FOREACH(__ifa, \
   1280 					    &(__ifp)->if_addrlist, ifa_list)
   1281 #define	IFADDR_FOREACH_SAFE(__ifa, __ifp, __nifa) \
   1282 					    TAILQ_FOREACH_SAFE(__ifa, \
   1283 					    &(__ifp)->if_addrlist, ifa_list, __nifa)
   1284 #define	IFADDR_EMPTY(__ifp)		TAILQ_EMPTY(&(__ifp)->if_addrlist)
   1285 
   1286 #define IFADDR_ENTRY_INIT(__ifa)					\
   1287 	PSLIST_ENTRY_INIT((__ifa), ifa_pslist_entry)
   1288 #define IFADDR_ENTRY_DESTROY(__ifa)					\
   1289 	PSLIST_ENTRY_DESTROY((__ifa), ifa_pslist_entry)
   1290 #define IFADDR_READER_EMPTY(__ifp)					\
   1291 	(PSLIST_READER_FIRST(&(__ifp)->if_addr_pslist, struct ifaddr,	\
   1292 	                     ifa_pslist_entry) == NULL)
   1293 #define IFADDR_READER_FIRST(__ifp)					\
   1294 	PSLIST_READER_FIRST(&(__ifp)->if_addr_pslist, struct ifaddr,	\
   1295 	                    ifa_pslist_entry)
   1296 #define IFADDR_READER_NEXT(__ifa)					\
   1297 	PSLIST_READER_NEXT((__ifa), struct ifaddr, ifa_pslist_entry)
   1298 #define IFADDR_READER_FOREACH(__ifa, __ifp)				\
   1299 	PSLIST_READER_FOREACH((__ifa), &(__ifp)->if_addr_pslist, struct ifaddr,\
   1300 	                      ifa_pslist_entry)
   1301 #define IFADDR_WRITER_INSERT_HEAD(__ifp, __ifa)				\
   1302 	PSLIST_WRITER_INSERT_HEAD(&(__ifp)->if_addr_pslist, (__ifa),	\
   1303 	                          ifa_pslist_entry)
   1304 #define IFADDR_WRITER_REMOVE(__ifa)					\
   1305 	PSLIST_WRITER_REMOVE((__ifa), ifa_pslist_entry)
   1306 #define IFADDR_WRITER_FOREACH(__ifa, __ifp)				\
   1307 	PSLIST_WRITER_FOREACH((__ifa), &(__ifp)->if_addr_pslist, struct ifaddr,\
   1308 	                      ifa_pslist_entry)
   1309 #define IFADDR_WRITER_NEXT(__ifp)					\
   1310 	PSLIST_WRITER_NEXT((__ifp), struct ifaddr, ifa_pslist_entry)
   1311 #define IFADDR_WRITER_INSERT_AFTER(__ifp, __new)			\
   1312 	PSLIST_WRITER_INSERT_AFTER((__ifp), (__new), ifa_pslist_entry)
   1313 #define IFADDR_WRITER_EMPTY(__ifp)					\
   1314 	(PSLIST_WRITER_FIRST(&(__ifp)->if_addr_pslist, struct ifaddr,	\
   1315 	                     ifa_pslist_entry) == NULL)
   1316 #define IFADDR_WRITER_INSERT_TAIL(__ifp, __new)				\
   1317 	do {								\
   1318 		if (IFADDR_WRITER_EMPTY(__ifp)) {			\
   1319 			IFADDR_WRITER_INSERT_HEAD((__ifp), (__new));	\
   1320 		} else {						\
   1321 			struct ifaddr *__ifa;				\
   1322 			IFADDR_WRITER_FOREACH(__ifa, (__ifp)) {		\
   1323 				if (IFADDR_WRITER_NEXT(__ifa) == NULL) {\
   1324 					IFADDR_WRITER_INSERT_AFTER(__ifa,\
   1325 					    (__new));			\
   1326 					break;				\
   1327 				}					\
   1328 			}						\
   1329 		}							\
   1330 	} while (0)
   1331 
   1332 #define	IFNET_GLOBAL_LOCK()			mutex_enter(&ifnet_mtx)
   1333 #define	IFNET_GLOBAL_UNLOCK()			mutex_exit(&ifnet_mtx)
   1334 #define	IFNET_GLOBAL_LOCKED()			mutex_owned(&ifnet_mtx)
   1335 
   1336 #define IFNET_READER_EMPTY() \
   1337 	(PSLIST_READER_FIRST(&ifnet_pslist, struct ifnet, if_pslist_entry) == NULL)
   1338 #define IFNET_READER_FIRST() \
   1339 	PSLIST_READER_FIRST(&ifnet_pslist, struct ifnet, if_pslist_entry)
   1340 #define IFNET_READER_NEXT(__ifp) \
   1341 	PSLIST_READER_NEXT((__ifp), struct ifnet, if_pslist_entry)
   1342 #define IFNET_READER_FOREACH(__ifp) \
   1343 	PSLIST_READER_FOREACH((__ifp), &ifnet_pslist, struct ifnet, \
   1344 	                      if_pslist_entry)
   1345 #define IFNET_WRITER_INSERT_HEAD(__ifp) \
   1346 	PSLIST_WRITER_INSERT_HEAD(&ifnet_pslist, (__ifp), if_pslist_entry)
   1347 #define IFNET_WRITER_REMOVE(__ifp) \
   1348 	PSLIST_WRITER_REMOVE((__ifp), if_pslist_entry)
   1349 #define IFNET_WRITER_FOREACH(__ifp) \
   1350 	PSLIST_WRITER_FOREACH((__ifp), &ifnet_pslist, struct ifnet, \
   1351 	                      if_pslist_entry)
   1352 #define IFNET_WRITER_NEXT(__ifp) \
   1353 	PSLIST_WRITER_NEXT((__ifp), struct ifnet, if_pslist_entry)
   1354 #define IFNET_WRITER_INSERT_AFTER(__ifp, __new) \
   1355 	PSLIST_WRITER_INSERT_AFTER((__ifp), (__new), if_pslist_entry)
   1356 #define IFNET_WRITER_EMPTY() \
   1357 	(PSLIST_WRITER_FIRST(&ifnet_pslist, struct ifnet, if_pslist_entry) == NULL)
   1358 #define IFNET_WRITER_INSERT_TAIL(__new)					\
   1359 	do {								\
   1360 		if (IFNET_WRITER_EMPTY()) {				\
   1361 			IFNET_WRITER_INSERT_HEAD(__new);		\
   1362 		} else {						\
   1363 			struct ifnet *__ifp;				\
   1364 			IFNET_WRITER_FOREACH(__ifp) {			\
   1365 				if (IFNET_WRITER_NEXT(__ifp) == NULL) {	\
   1366 					IFNET_WRITER_INSERT_AFTER(__ifp,\
   1367 					    (__new));			\
   1368 					break;				\
   1369 				}					\
   1370 			}						\
   1371 		}							\
   1372 	} while (0)
   1373 
   1374 #define IFNET_LOCK(ifp)		mutex_enter((ifp)->if_ioctl_lock)
   1375 #define IFNET_UNLOCK(ifp)	mutex_exit((ifp)->if_ioctl_lock)
   1376 #define IFNET_LOCKED(ifp)	mutex_owned((ifp)->if_ioctl_lock)
   1377 
   1378 #define IFNET_ASSERT_UNLOCKED(ifp)	\
   1379 	KDASSERT(mutex_ownable((ifp)->if_ioctl_lock))
   1380 
   1381 extern struct pslist_head ifnet_pslist;
   1382 extern kmutex_t ifnet_mtx;
   1383 
   1384 extern struct ifnet *lo0ifp;
   1385 
   1386 /*
   1387  * ifq sysctl support
   1388  */
   1389 int	sysctl_ifq(int *name, u_int namelen, void *oldp,
   1390 		       size_t *oldlenp, void *newp, size_t newlen,
   1391 		       struct ifqueue *ifq);
   1392 /* symbolic names for terminal (per-protocol) CTL_IFQ_ nodes */
   1393 #define IFQCTL_LEN	1
   1394 #define IFQCTL_MAXLEN	2
   1395 #define IFQCTL_PEAK	3
   1396 #define IFQCTL_DROPS	4
   1397 
   1398 /*
   1399  * Hook for if_vlan - needed by if_agr
   1400  */
   1401 MODULE_HOOK(if_vlan_vlan_input_hook, void, (struct ifnet *, struct mbuf *));
   1402 
   1403 #endif /* _KERNEL */
   1404 
   1405 #endif /* !_NET_IF_H_ */
   1406