| blob | 833c4632c01fee0a39385c5c198c5d79f7fa75bd |
1 .\" Copyright (c) 1983, 1991, 1993
2 .\" The Regents of the University of California. All rights reserved.
3 .\"
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
6 .\" are met:
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
12 .\" 3. All advertising materials mentioning features or use of this software
13 .\" must display the following acknowledgement:
14 .\" This product includes software developed by the University of
15 .\" California, Berkeley and its contributors.
16 .\" 4. Neither the name of the University nor the names of its contributors
17 .\" may be used to endorse or promote products derived from this software
18 .\" without specific prior written permission.
19 .\"
20 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30 .\" SUCH DAMAGE.
31 .\"
32 .\" @(#)ip.4 8.2 (Berkeley) 11/30/93
33 .\" $FreeBSD$
34 .\"
35 .Dd April 9, 2007
36 .Dt IP 4
37 .Os
38 .Sh NAME
39 .Nm ip
40 .Nd Internet Protocol
41 .Sh SYNOPSIS
42 .In sys/types.h
43 .In sys/socket.h
44 .In netinet/in.h
45 .Ft int
46 .Fn socket AF_INET SOCK_RAW proto
47 .Sh DESCRIPTION
48 .Tn IP
49 is the transport layer protocol used
50 by the Internet protocol family.
51 Options may be set at the
52 .Tn IP
53 level
54 when using higher-level protocols that are based on
55 .Tn IP
56 (such as
57 .Tn TCP
58 and
59 .Tn UDP ) .
60 It may also be accessed
61 through a
62 .Dq raw socket
63 when developing new protocols, or
64 special-purpose applications.
65 .Pp
66 There are several
67 .Tn IP-level
68 .Xr setsockopt 2
69 and
70 .Xr getsockopt 2
71 options.
72 .Dv IP_OPTIONS
73 may be used to provide
74 .Tn IP
75 options to be transmitted in the
76 .Tn IP
77 header of each outgoing packet
78 or to examine the header options on incoming packets.
79 .Tn IP
80 options may be used with any socket type in the Internet family.
81 The format of
82 .Tn IP
83 options to be sent is that specified by the
84 .Tn IP
85 protocol specification (RFC-791), with one exception:
86 the list of addresses for Source Route options must include the first-hop
87 gateway at the beginning of the list of gateways.
88 The first-hop gateway address will be extracted from the option list
89 and the size adjusted accordingly before use.
90 To disable previously specified options,
91 use a zero-length buffer:
92 .Bd -literal
93 setsockopt(s, IPPROTO_IP, IP_OPTIONS, NULL, 0);
94 .Ed
95 .Pp
96 .Dv IP_TOS
97 and
98 .Dv IP_TTL
99 may be used to set the type-of-service and time-to-live
100 fields in the
101 .Tn IP
102 header for
103 .Dv SOCK_STREAM , SOCK_DGRAM ,
104 and certain types of
105 .Dv SOCK_RAW
106 sockets.
107 For example,
108 .Bd -literal
109 int tos = IPTOS_LOWDELAY; /* see <netinet/ip.h> */
110 setsockopt(s, IPPROTO_IP, IP_TOS, &tos, sizeof(tos));
112 int ttl = 60; /* max = 255 */
113 setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(ttl));
114 .Ed
115 .Pp
116 .Dv IP_MINTTL
117 may be used to set the minimum acceptable TTL a packet must have when
118 received on a socket.
119 All packets with a lower TTL are silently dropped.
120 This option is only really useful when set to 255, preventing packets
121 from outside the directly connected networks reaching local listeners
122 on sockets.
123 .Pp
124 .Dv IP_DONTFRAG
125 may be used to set the Don't Fragment flag on IP packets.
126 Currently this option is respected only on
127 .Xr udp 4
128 and raw
129 .Xr ip 4
130 sockets, unless the
131 .Dv IP_HDRINCL
132 option has been set.
133 On
134 .Xr tcp 4
135 sockets, the Don't Fragment flag is controlled by the Path
136 MTU Discovery option.
137 Sending a packet larger than the MTU size of the egress interface,
138 determined by the destination address, returns an
139 .Er EMSGSIZE
140 error.
141 .Pp
142 If the
143 .Dv IP_RECVDSTADDR
144 option is enabled on a
145 .Dv SOCK_DGRAM
146 socket,
147 the
148 .Xr recvmsg 2
149 call will return the destination
150 .Tn IP
151 address for a
152 .Tn UDP
153 datagram.
154 The
155 .Vt msg_control
156 field in the
157 .Vt msghdr
158 structure points to a buffer
159 that contains a
160 .Vt cmsghdr
161 structure followed by the
162 .Tn IP
163 address.
164 The
165 .Vt cmsghdr
166 fields have the following values:
167 .Bd -literal
168 cmsg_len = sizeof(struct in_addr)
169 cmsg_level = IPPROTO_IP
170 cmsg_type = IP_RECVDSTADDR
171 .Ed
172 .Pp
173 The source address to be used for outgoing
174 .Tn UDP
175 datagrams on a socket that is not bound to a specific
176 .Tn IP
177 address can be specified as ancillary data with a type code of
178 .Dv IP_SENDSRCADDR .
179 The msg_control field in the msghdr structure should point to a buffer
180 that contains a
181 .Vt cmsghdr
182 structure followed by the
183 .Tn IP
184 address.
185 The cmsghdr fields should have the following values:
186 .Bd -literal
187 cmsg_len = sizeof(struct in_addr)
188 cmsg_level = IPPROTO_IP
189 cmsg_type = IP_SENDSRCADDR
190 .Ed
191 .Pp
192 For convenience,
193 .Dv IP_SENDSRCADDR
194 is defined to have the same value as
195 .Dv IP_RECVDSTADDR ,
196 so the
197 .Dv IP_RECVDSTADDR
198 control message from
199 .Xr recvmsg 2
200 can be used directly as a control message for
201 .Xr sendmsg 2 .
202 .\"
203 .Pp
204 If the
205 .Dv IP_ONESBCAST
206 option is enabled on a
207 .Dv SOCK_DGRAM
208 or a
209 .Dv SOCK_RAW
210 socket, the destination address of outgoing
211 broadcast datagrams on that socket will be forced
212 to the undirected broadcast address,
213 .Dv INADDR_BROADCAST ,
214 before transmission.
215 This is in contrast to the default behavior of the
216 system, which is to transmit undirected broadcasts
217 via the first network interface with the
218 .Dv IFF_BROADCAST flag set.
219 .Pp
220 This option allows applications to choose which
221 interface is used to transmit an undirected broadcast
222 datagram.
223 For example, the following code would force an
224 undirected broadcast to be transmitted via the interface
225 configured with the broadcast address 192.168.2.255:
226 .Bd -literal
227 char msg[512];
228 struct sockaddr_in sin;
229 u_char onesbcast = 1; /* 0 = disable (default), 1 = enable */
231 setsockopt(s, IPPROTO_IP, IP_ONESBCAST, &onesbcast, sizeof(onesbcast));
232 sin.sin_addr.s_addr = inet_addr("192.168.2.255");
233 sin.sin_port = htons(1234);
234 sendto(s, msg, sizeof(msg), 0, &sin, sizeof(sin));
235 .Ed
236 .Pp
237 It is the application's responsibility to set the
238 .Dv IP_TTL option
239 to an appropriate value in order to prevent broadcast storms.
240 The application must have sufficient credentials to set the
241 .Dv SO_BROADCAST
242 socket level option, otherwise the
243 .Dv IP_ONESBCAST option has no effect.
244 .Pp
245 If the
246 .Dv IP_RECVTTL
247 option is enabled on a
248 .Dv SOCK_DGRAM
249 socket, the
250 .Xr recvmsg 2
251 call will return the
252 .Tn IP
253 .Tn TTL
254 (time to live) field for a
255 .Tn UDP
256 datagram.
257 The msg_control field in the msghdr structure points to a buffer
258 that contains a cmsghdr structure followed by the
259 .Tn TTL .
260 The cmsghdr fields have the following values:
261 .Bd -literal
262 cmsg_len = sizeof(u_char)
263 cmsg_level = IPPROTO_IP
264 cmsg_type = IP_RECVTTL
265 .Ed
266 .\"
267 .Pp
268 If the
269 .Dv IP_RECVIF
270 option is enabled on a
271 .Dv SOCK_DGRAM
272 socket, the
273 .Xr recvmsg 2
274 call returns a
275 .Vt "struct sockaddr_dl"
276 corresponding to the interface on which the
277 packet was received.
278 The
279 .Va msg_control
280 field in the
281 .Vt msghdr
282 structure points to a buffer that contains a
283 .Vt cmsghdr
284 structure followed by the
285 .Vt "struct sockaddr_dl" .
286 The
287 .Vt cmsghdr
288 fields have the following values:
289 .Bd -literal
290 cmsg_len = sizeof(struct sockaddr_dl)
291 cmsg_level = IPPROTO_IP
292 cmsg_type = IP_RECVIF
293 .Ed
294 .Pp
295 .Dv IP_PORTRANGE
296 may be used to set the port range used for selecting a local port number
297 on a socket with an unspecified (zero) port number.
298 It has the following
299 possible values:
300 .Bl -tag -width IP_PORTRANGE_DEFAULT
301 .It Dv IP_PORTRANGE_DEFAULT
302 use the default range of values, normally
303 .Dv IPPORT_HIFIRSTAUTO
304 through
305 .Dv IPPORT_HILASTAUTO .
306 This is adjustable through the sysctl setting:
307 .Va net.inet.ip.portrange.first
308 and
309 .Va net.inet.ip.portrange.last .
310 .It Dv IP_PORTRANGE_HIGH
311 use a high range of values, normally
312 .Dv IPPORT_HIFIRSTAUTO
313 and
314 .Dv IPPORT_HILASTAUTO .
315 This is adjustable through the sysctl setting:
316 .Va net.inet.ip.portrange.hifirst
317 and
318 .Va net.inet.ip.portrange.hilast .
319 .It Dv IP_PORTRANGE_LOW
320 use a low range of ports, which are normally restricted to
321 privileged processes on
322 .Ux
323 systems.
324 The range is normally from
325 .Dv IPPORT_RESERVED
326 \- 1 down to
327 .Li IPPORT_RESERVEDSTART
328 in descending order.
329 This is adjustable through the sysctl setting:
330 .Va net.inet.ip.portrange.lowfirst
331 and
332 .Va net.inet.ip.portrange.lowlast .
333 .El
334 .Pp
335 The range of privileged ports which only may be opened by
336 root-owned processes may be modified by the
337 .Va net.inet.ip.portrange.reservedlow
338 and
339 .Va net.inet.ip.portrange.reservedhigh
340 sysctl settings.
341 The values default to the traditional range,
342 0 through
343 .Dv IPPORT_RESERVED
344 \- 1
345 (0 through 1023), respectively.
346 Note that these settings do not affect and are not accounted for in the
347 use or calculation of the other
348 .Va net.inet.ip.portrange
349 values above.
350 Changing these values departs from
351 .Ux
352 tradition and has security
353 consequences that the administrator should carefully evaluate before
354 modifying these settings.
355 .Pp
356 Ports are allocated at random within the specified port range in order
357 to increase the difficulty of random spoofing attacks.
358 In scenarios such as benchmarking, this behavior may be undesirable.
359 In these cases,
360 .Va net.inet.ip.portrange.randomized
361 can be used to toggle randomization off.
362 If more than
363 .Va net.inet.ip.portrange.randomcps
364 ports have been allocated in the last second, then return to sequential
365 port allocation.
366 Return to random allocation only once the current port allocation rate
367 drops below
368 .Va net.inet.ip.portrange.randomcps
369 for at least
370 .Va net.inet.ip.portrange.randomtime
371 seconds.
372 The default values for
373 .Va net.inet.ip.portrange.randomcps
374 and
375 .Va net.inet.ip.portrange.randomtime
376 are 10 port allocations per second and 45 seconds correspondingly.
377 .Ss "Multicast Options"
378 .Pp
379 .Tn IP
380 multicasting is supported only on
381 .Dv AF_INET
382 sockets of type
383 .Dv SOCK_DGRAM
384 and
385 .Dv SOCK_RAW ,
386 and only on networks where the interface
387 driver supports multicasting.
388 .Pp
389 The
390 .Dv IP_MULTICAST_TTL
391 option changes the time-to-live (TTL)
392 for outgoing multicast datagrams
393 in order to control the scope of the multicasts:
394 .Bd -literal
395 u_char ttl; /* range: 0 to 255, default = 1 */
396 setsockopt(s, IPPROTO_IP, IP_MULTICAST_TTL, &ttl, sizeof(ttl));
397 .Ed
398 .Pp
399 Datagrams with a TTL of 1 are not forwarded beyond the local network.
400 Multicast datagrams with a TTL of 0 will not be transmitted on any network,
401 but may be delivered locally if the sending host belongs to the destination
402 group and if multicast loopback has not been disabled on the sending socket
403 (see below).
404 Multicast datagrams with TTL greater than 1 may be forwarded
405 to other networks if a multicast router is attached to the local network.
406 .Pp
407 For hosts with multiple interfaces, each multicast transmission is
408 sent from the primary network interface.
409 The
410 .Dv IP_MULTICAST_IF
411 option overrides the default for
412 subsequent transmissions from a given socket:
413 .Bd -literal
414 struct in_addr addr;
415 setsockopt(s, IPPROTO_IP, IP_MULTICAST_IF, &addr, sizeof(addr));
416 .Ed
417 .Pp
418 where "addr" is the local
419 .Tn IP
420 address of the desired interface or
421 .Dv INADDR_ANY
422 to specify the default interface.
423 .Pp
424 To specify an interface by index, an instance of
425 .Vt ip_mreqn
426 should be passed instead.
427 The
428 .Vt imr_ifindex
429 member should be set to the index of the desired interface,
430 or 0 to specify the default interface.
431 The kernel differentiates between these two structures by their size.
432 .\"
433 An interface's local IP address and multicast capability can
434 be obtained via the
435 .Dv SIOCGIFCONF
436 and
437 .Dv SIOCGIFFLAGS
438 ioctls.
439 Normal applications should not need to use this option.
440 .Pp
441 If a multicast datagram is sent to a group to which the sending host itself
442 belongs (on the outgoing interface), a copy of the datagram is, by default,
443 looped back by the IP layer for local delivery.
444 The
445 .Dv IP_MULTICAST_LOOP
446 option gives the sender explicit control
447 over whether or not subsequent datagrams are looped back:
448 .Bd -literal
449 u_char loop; /* 0 = disable, 1 = enable (default) */
450 setsockopt(s, IPPROTO_IP, IP_MULTICAST_LOOP, &loop, sizeof(loop));
451 .Ed
452 .Pp
453 This option
454 improves performance for applications that may have no more than one
455 instance on a single host (such as a router daemon), by eliminating
456 the overhead of receiving their own transmissions.
457 It should generally not
458 be used by applications for which there may be more than one instance on a
459 single host (such as a conferencing program) or for which the sender does
460 not belong to the destination group (such as a time querying program).
461 .Pp
462 A multicast datagram sent with an initial TTL greater than 1 may be delivered
463 to the sending host on a different interface from that on which it was sent,
464 if the host belongs to the destination group on that other interface.
465 The loopback control option has no effect on such delivery.
466 .Pp
467 A host must become a member of a multicast group before it can receive
468 datagrams sent to the group.
469 To join a multicast group, use the
470 .Dv IP_ADD_MEMBERSHIP
471 option:
472 .Bd -literal
473 struct ip_mreq mreq;
474 setsockopt(s, IPPROTO_IP, IP_ADD_MEMBERSHIP, &mreq, sizeof(mreq));
475 .Ed
476 .Pp
477 where
478 .Fa mreq
479 is the following structure:
480 .Bd -literal
481 struct ip_mreq {
482 struct in_addr imr_multiaddr; /* IP multicast address of group */
483 struct in_addr imr_interface; /* local IP address of interface */
484 }
485 .Ed
486 .Pp
487 .Va imr_interface
488 should be set to
489 .Dv INADDR_ANY
490 to choose the default multicast interface,
491 or the
492 .Tn IP
493 address of a particular multicast-capable interface if
494 the host is multihomed.
495 .\" TODO: Remove this piece when the RFC 3678 API is implemented and
496 .\" the RFC 1724 hack is removed.
497 Since
498 .Fx 4.4 ,
499 if the
500 .Va imr_interface
501 member is within the network range
502 .Li 0.0.0.0/8 ,
503 it is treated as an interface index in the system interface MIB,
504 as per the RIP Version 2 MIB Extension (RFC-1724).
505 .\" TODO: Update this piece when IPv4 source-address selection is implemented.
506 .Pp
507 Up to
508 .Dv IP_MAX_MEMBERSHIPS
509 memberships may be added on a single socket.
510 Membership is associated with a single interface;
511 programs running on multihomed hosts may need to
512 join the same group on more than one interface.
513 .Pp
514 The IGMP protocol uses the primary IP address of the interface
515 as its identifier for group membership.
516 If multiple IP aliases are configured on the same interface,
517 they will be ignored.
518 This shortcoming was addressed in IPv6; MLDv2 requires
519 that the unique link-local address for an interface is
520 used to identify an MLDv2 listener.
521 .Pp
522 To drop a membership, use:
523 .Bd -literal
524 struct ip_mreq mreq;
525 setsockopt(s, IPPROTO_IP, IP_DROP_MEMBERSHIP, &mreq, sizeof(mreq));
526 .Ed
527 .Pp
528 where
529 .Fa mreq
530 contains the same values as used to add the membership.
531 Memberships are dropped when the socket is closed or the process exits.
532 .\"-----------------------
533 .Ss "Raw IP Sockets"
534 .Pp
535 Raw
536 .Tn IP
537 sockets are connectionless,
538 and are normally used with the
539 .Xr sendto 2
540 and
541 .Xr recvfrom 2
542 calls, though the
543 .Xr connect 2
544 call may also be used to fix the destination for future
545 packets (in which case the
546 .Xr read 2
547 or
548 .Xr recv 2
549 and
550 .Xr write 2
551 or
552 .Xr send 2
553 system calls may be used).
554 .Pp
555 If
556 .Fa proto
557 is 0, the default protocol
558 .Dv IPPROTO_RAW
559 is used for outgoing
560 packets, and only incoming packets destined for that protocol
561 are received.
562 If
563 .Fa proto
564 is non-zero, that protocol number will be used on outgoing packets
565 and to filter incoming packets.
566 .Pp
567 Outgoing packets automatically have an
568 .Tn IP
569 header prepended to
570 them (based on the destination address and the protocol
571 number the socket is created with),
572 unless the
573 .Dv IP_HDRINCL
574 option has been set.
575 Incoming packets are received with
576 .Tn IP
577 header and options intact.
578 .Pp
579 .Dv IP_HDRINCL
580 indicates the complete IP header is included with the data
581 and may be used only with the
582 .Dv SOCK_RAW
583 type.
584 .Bd -literal
585 #include <netinet/in_systm.h>
586 #include <netinet/ip.h>
588 int hincl = 1; /* 1 = on, 0 = off */
589 setsockopt(s, IPPROTO_IP, IP_HDRINCL, &hincl, sizeof(hincl));
590 .Ed
591 .Pp
592 Unlike previous
593 .Bx
594 releases, the program must set all
595 the fields of the IP header, including the following:
596 .Bd -literal
597 ip->ip_v = IPVERSION;
598 ip->ip_hl = hlen >> 2;
599 ip->ip_id = 0; /* 0 means kernel set appropriate value */
600 ip->ip_off = offset;
601 .Ed
602 .Pp
603 The
604 .Va ip_len
605 and
606 .Va ip_off
607 fields
608 .Em must
609 be provided in host byte order .
610 All other fields must be provided in network byte order.
611 See
612 .Xr byteorder 3
613 for more information on network byte order.
614 If the
615 .Va ip_id
616 field is set to 0 then the kernel will choose an
617 appropriate value.
618 If the header source address is set to
619 .Dv INADDR_ANY ,
620 the kernel will choose an appropriate address.
621 .Sh ERRORS
622 A socket operation may fail with one of the following errors returned:
623 .Bl -tag -width Er
624 .It Bq Er EISCONN
625 when trying to establish a connection on a socket which
626 already has one, or when trying to send a datagram with the destination
627 address specified and the socket is already connected;
628 .It Bq Er ENOTCONN
629 when trying to send a datagram, but
630 no destination address is specified, and the socket has not been
631 connected;
632 .It Bq Er ENOBUFS
633 when the system runs out of memory for
634 an internal data structure;
635 .It Bq Er EADDRNOTAVAIL
636 when an attempt is made to create a
637 socket with a network address for which no network interface
638 exists.
639 .It Bq Er EACCES
640 when an attempt is made to create
641 a raw IP socket by a non-privileged process.
642 .El
643 .Pp
644 The following errors specific to
645 .Tn IP
646 may occur when setting or getting
647 .Tn IP
648 options:
649 .Bl -tag -width Er
650 .It Bq Er EINVAL
651 An unknown socket option name was given.
652 .It Bq Er EINVAL
653 The IP option field was improperly formed;
654 an option field was shorter than the minimum value
655 or longer than the option buffer provided.
656 .El
657 .Pp
658 The following errors may occur when attempting to send
659 .Tn IP
660 datagrams via a
661 .Dq raw socket
662 with the
663 .Dv IP_HDRINCL
664 option set:
665 .Bl -tag -width Er
666 .It Bq Er EINVAL
667 The user-supplied
668 .Va ip_len
669 field was not equal to the length of the datagram written to the socket.
670 .El
671 .Sh SEE ALSO
672 .Xr getsockopt 2 ,
673 .Xr recv 2 ,
674 .Xr send 2 ,
675 .Xr byteorder 3 ,
676 .Xr icmp 4 ,
677 .Xr inet 4 ,
678 .Xr intro 4 ,
679 .Xr multicast 4
680 .Sh HISTORY
681 The
682 .Nm
683 protocol appeared in
684 .Bx 4.2 .
685 The
686 .Vt ip_mreqn
687 structure appeared in
688 .Tn Linux 2.4 .