xref: /openbsd-src/share/man/man4/ip.4 (revision b2ea75c1b17e1a9a339660e7ed45cd24946b230e)
1.\"	$OpenBSD: ip.4,v 1.15 2001/05/14 09:12:06 markus Exp $
2.\"	$NetBSD: ip.4,v 1.3 1994/11/30 16:22:19 jtc Exp $
3.\"
4.\" Copyright (c) 1983, 1991, 1993
5.\"	The Regents of the University of California.  All rights reserved.
6.\"
7.\" Redistribution and use in source and binary forms, with or without
8.\" modification, are permitted provided that the following conditions
9.\" are met:
10.\" 1. Redistributions of source code must retain the above copyright
11.\"    notice, this list of conditions and the following disclaimer.
12.\" 2. Redistributions in binary form must reproduce the above copyright
13.\"    notice, this list of conditions and the following disclaimer in the
14.\"    documentation and/or other materials provided with the distribution.
15.\" 3. All advertising materials mentioning features or use of this software
16.\"    must display the following acknowledgement:
17.\"	This product includes software developed by the University of
18.\"	California, Berkeley and its contributors.
19.\" 4. Neither the name of the University nor the names of its contributors
20.\"    may be used to endorse or promote products derived from this software
21.\"    without specific prior written permission.
22.\"
23.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
24.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
25.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
26.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
27.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
28.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
29.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
30.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
31.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
32.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33.\" SUCH DAMAGE.
34.\"
35.\"     @(#)ip.4	8.2 (Berkeley) 11/30/93
36.\"
37.Dd November 30, 1993
38.Dt IP 4
39.Os
40.Sh NAME
41.Nm ip
42.Nd Internet Protocol
43.Sh SYNOPSIS
44.Fd #include <sys/socket.h>
45.Fd #include <netinet/in.h>
46.Ft int
47.Fn socket AF_INET SOCK_RAW proto
48.Sh DESCRIPTION
49.Tn IP
50is the network layer protocol used
51by the Internet protocol family.
52Options may be set at the
53.Tn IP
54level
55when using higher-level protocols that are based on
56.Tn IP
57(such as
58.Tn TCP
59and
60.Tn UDP ) .
61It may also be accessed
62through a
63.Dq raw socket
64when developing new protocols, or
65special-purpose applications.
66.Pp
67There are several
68.Tn IP-level
69.Xr setsockopt 2 Ns / Ns Xr getsockopt 2
70options.
71.Dv IP_OPTIONS
72may be used to provide
73.Tn IP
74options to be transmitted in the
75.Tn IP
76header of each outgoing packet
77or to examine the header options on incoming packets.
78.Tn IP
79options may be used with any socket type in the Internet family.
80The format of
81.Tn IP
82options to be sent is that specified by the
83.Tn IP
84protocol specification (RFC-791), with one exception:
85the list of addresses for Source Route options must include the first-hop
86gateway at the beginning of the list of gateways.
87The first-hop gateway address will be extracted from the option list
88and the size adjusted accordingly before use.
89To disable previously specified options,
90use a zero-length buffer:
91.Bd -literal
92setsockopt(s, IPPROTO_IP, IP_OPTIONS, NULL, 0);
93.Ed
94.Pp
95.Dv IP_TOS
96and
97.Dv IP_TTL
98may be used to set the type-of-service and time-to-live
99fields in the
100.Tn IP
101header for
102.Dv SOCK_STREAM
103and
104.Dv SOCK_DGRAM
105sockets. For example,
106.Bd -literal
107int tos = IPTOS_LOWDELAY;       /* see <netinet/in.h> */
108setsockopt(s, IPPROTO_IP, IP_TOS, &tos, sizeof(tos));
109
110int ttl = 60;                   /* max = 255 */
111setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(ttl));
112.Ed
113.Pp
114If the
115.Dv IP_RECVDSTADDR
116option is enabled on a
117.Dv SOCK_DGRAM
118socket,
119the
120.Xr recvmsg
121call will return the destination
122.Tn IP
123address for a
124.Tn UDP
125datagram.
126The msg_control field in the msghdr structure points to a buffer
127that contains a cmsghdr structure followed by the
128.Tn IP
129address.
130The cmsghdr fields have the following values:
131.Bd -literal
132cmsg_len = CMSG_LEN(sizeof(struct in_addr))
133cmsg_level = IPPROTO_IP
134cmsg_type = IP_RECVDSTADDR
135.Ed
136.Pp
137The IP_PORTRANGE
138option causes the default allocation policy for when the kernel is asked
139to choose a free port number.  Three choices are available:
140.Bl -tag -width IP_PORTRANGE_DEFAULT
141.It IP_PORTRANGE_DEFAULT
142The regular range of non-reserved ports.
143.It IP_PORTRANGE_HIGH
144A high range, for fun.
145.It IP_PORTRANGE_LOW
146Reserved ports; between 600 and 1023.
147.El
148.Ss "Multicast Options"
149.Pp
150.Tn IP
151multicasting is supported only on
152.Dv AF_INET
153sockets of type
154.Dv SOCK_DGRAM
155and
156.Dv SOCK_RAW,
157and only on networks where the interface
158driver supports multicasting.
159.Pp
160The
161.Dv IP_MULTICAST_TTL
162option changes the time-to-live (TTL)
163for outgoing multicast datagrams
164in order to control the scope of the multicasts:
165.Bd -literal
166u_char ttl;	/* range: 0 to 255, default = 1 */
167setsockopt(s, IPPROTO_IP, IP_MULTICAST_TTL, &ttl, sizeof(ttl));
168.Ed
169.sp
170Datagrams with a TTL of 1 are not forwarded beyond the local network.
171Multicast datagrams with a TTL of 0 will not be transmitted on any network,
172but may be delivered locally if the sending host belongs to the destination
173group and if multicast loopback has not been disabled on the sending socket
174(see below).  Multicast datagrams with TTL greater than 1 may be forwarded
175to other networks if a multicast router is attached to the local network.
176.Pp
177For hosts with multiple interfaces, each multicast transmission is
178sent from the primary network interface.
179The
180.Dv IP_MULTICAST_IF
181option overrides the default for
182subsequent transmissions from a given socket:
183.Bd -literal
184struct in_addr addr;
185setsockopt(s, IPPROTO_IP, IP_MULTICAST_IF, &addr, sizeof(addr));
186.Ed
187.sp
188where "addr" is the local
189.Tn IP
190address of the desired interface or
191.Dv INADDR_ANY
192to specify the default interface.
193An interface's local IP address and multicast capability can
194be obtained via the
195.Dv SIOCGIFCONF
196and
197.Dv SIOCGIFFLAGS
198ioctls.
199Normal applications should not need to use this option.
200.Pp
201If a multicast datagram is sent to a group to which the sending host itself
202belongs (on the outgoing interface), a copy of the datagram is, by default,
203looped back by the IP layer for local delivery.
204The
205.Dv IP_MULTICAST_LOOP
206option gives the sender explicit control
207over whether or not subsequent datagrams are looped back:
208.Bd -literal
209u_char loop;	/* 0 = disable, 1 = enable (default) */
210setsockopt(s, IPPROTO_IP, IP_MULTICAST_LOOP, &loop, sizeof(loop));
211.Ed
212.sp
213This option
214improves performance for applications that may have no more than one
215instance on a single host (such as a router daemon), by eliminating
216the overhead of receiving their own transmissions.  It should generally not
217be used by applications for which there may be more than one instance on a
218single host (such as a conferencing program) or for which the sender does
219not belong to the destination group (such as a time querying program).
220.Pp
221A multicast datagram sent with an initial TTL greater than 1 may be delivered
222to the sending host on a different interface from that on which it was sent,
223if the host belongs to the destination group on that other interface.  The
224loopback control option has no effect on such delivery.
225.Pp
226A host must become a member of a multicast group before it can receive
227datagrams sent to the group.  To join a multicast group, use the
228.Dv IP_ADD_MEMBERSHIP
229option:
230.Bd -literal
231struct ip_mreq mreq;
232setsockopt(s, IPPROTO_IP, IP_ADD_MEMBERSHIP, &mreq, sizeof(mreq));
233.Ed
234.sp
235where
236.Fa mreq
237is the following structure:
238.Bd -literal
239struct ip_mreq {
240    struct in_addr imr_multiaddr; /* multicast group to join */
241    struct in_addr imr_interface; /* interface to join on */
242}
243.Ed
244.sp
245.Dv imr_interface
246should
247be
248.Dv INADDR_ANY
249to choose the default multicast interface,
250or the
251.Tn IP
252address of a particular multicast-capable interface if
253the host is multihomed.
254Membership is associated with a single interface;
255programs running on multihomed hosts may need to
256join the same group on more than one interface.
257Up to
258.Dv IP_MAX_MEMBERSHIPS
259(currently 20) memberships may be added on a
260single socket.
261.Pp
262To drop a membership, use:
263.Bd -literal
264struct ip_mreq mreq;
265setsockopt(s, IPPROTO_IP, IP_DROP_MEMBERSHIP, &mreq, sizeof(mreq));
266.Ed
267.sp
268where
269.Fa mreq
270contains the same values as used to add the membership.
271Memberships are dropped when the socket is closed or the process exits.
272.\"-----------------------
273.Ss "Raw IP Sockets"
274.Pp
275Raw
276.Tn IP
277sockets are connectionless,
278and are normally used with the
279.Xr sendto
280and
281.Xr recvfrom
282calls, though the
283.Xr connect 2
284call may also be used to fix the destination for future
285packets (in which case the
286.Xr read 2
287or
288.Xr recv 2
289and
290.Xr write 2
291or
292.Xr send 2
293system calls may be used).
294.Pp
295If
296.Fa proto
297is 0, the default protocol
298.Dv IPPROTO_RAW
299is used for outgoing
300packets, and only incoming packets destined for that protocol
301are received.
302If
303.Fa proto
304is non-zero, that protocol number will be used on outgoing packets
305and to filter incoming packets.
306.Pp
307Outgoing packets automatically have an
308.Tn IP
309header prepended to
310them (based on the destination address and the protocol
311number the socket is created with),
312unless the
313.Dv IP_HDRINCL
314option has been set.
315Incoming packets are received with
316.Tn IP
317header and options intact.
318.Pp
319.Dv IP_HDRINCL
320indicates the complete IP header is included with the data
321and may be used only with the
322.Dv SOCK_RAW
323type.
324.Bd -literal
325#include <netinet/ip.h>
326
327int hincl = 1;                  /* 1 = on, 0 = off */
328setsockopt(s, IPPROTO_IP, IP_HDRINCL, &hincl, sizeof(hincl));
329.Ed
330.sp
331Unlike previous
332.Tn BSD
333releases, the program must set all
334the fields of the IP header, including the following:
335.Bd -literal
336ip->ip_v = IPVERSION;
337ip->ip_hl = hlen >> 2;
338ip->ip_id = 0;  /* 0 means kernel set appropriate value */
339ip->ip_off = htons(offset);
340ip->ip_len = htons(len);
341.Ed
342.sp .5
343.Pp
344Additionally note that starting with
345.Ox 2.1 ,
346the ip_off and ip_len fields are in network byte order.
347If the header source address is set to
348.Dv INADDR_ANY,
349the kernel will choose an appropriate address.
350.Sh DIAGNOSTICS
351A socket operation may fail with one of the following errors returned:
352.Bl -tag -width [EADDRNOTAVAIL]
353.It Bq Er EISCONN
354when trying to establish a connection on a socket which
355already has one, or when trying to send a datagram with the destination
356address specified and the socket is already connected;
357.It Bq Er ENOTCONN
358when trying to send a datagram, but
359no destination address is specified, and the socket hasn't been
360connected;
361.It Bq Er ENOBUFS
362when the system runs out of memory for
363an internal data structure;
364.It Bq Er EADDRNOTAVAIL
365when an attempt is made to create a
366socket with a network address for which no network interface
367exists.
368.It Bq Er EACCES
369when an attempt is made to create
370a raw IP socket by a non-privileged process.
371.El
372.Pp
373The following errors specific to
374.Tn IP
375may occur when setting or getting
376.Tn IP
377options:
378.Bl -tag -width EADDRNOTAVAILxx
379.It Bq Er EINVAL
380An unknown socket option name was given.
381.It Bq Er EINVAL
382The IP option field was improperly formed;
383an option field was shorter than the minimum value
384or longer than the option buffer provided.
385.El
386.Sh SEE ALSO
387.Xr getsockopt 2 ,
388.Xr recv 2 ,
389.Xr send 2 ,
390.Xr icmp 4 ,
391.Xr inet 4 ,
392.Xr netintro 4
393.Sh HISTORY
394The
395.Nm
396protocol appeared in
397.Bx 4.2 .
398