xref: /dragonfly/lib/libc/sys/socket.2 (revision 10f4bf95)
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.\"     From: @(#)socket.2	8.1 (Berkeley) 6/4/93
33.\" $FreeBSD: src/lib/libc/sys/socket.2,v 1.12.2.11 2002/12/29 16:35:34 schweikh Exp $
34.\" $DragonFly: src/lib/libc/sys/socket.2,v 1.5 2008/04/15 19:19:49 swildner Exp $
35.\"
36.Dd November 24, 1997
37.Dt SOCKET 2
38.Os
39.Sh NAME
40.Nm socket
41.Nd create an endpoint for communication
42.Sh LIBRARY
43.Lb libc
44.Sh SYNOPSIS
45.In sys/types.h
46.In sys/socket.h
47.Ft int
48.Fn socket "int domain" "int type" "int protocol"
49.Sh DESCRIPTION
50.Fn Socket
51creates an endpoint for communication and returns a descriptor.
52.Pp
53The
54.Fa domain
55parameter specifies a communications domain within which
56communication will take place; this selects the protocol family
57which should be used.
58These families are defined in the include file
59.In sys/socket.h .
60The currently understood formats are:
61.Pp
62.Bd -literal -offset indent -compact
63PF_LOCAL	Host-internal protocols, formerly called PF_UNIX,
64PF_UNIX		Host-internal protocols, deprecated, use PF_LOCAL,
65PF_INET		Internet version 4 protocols,
66PF_IMPLINK	ARPAnet IMP addresses,
67PF_PUP		PUP protocols, like BSP,
68PF_CHAOS	MIT CHAOS protocols,
69PF_NS		Xerox Network Systems protocols,
70PF_ISO		ISO protocols,
71PF_OSI		Open Systems Interconnection protocols,
72PF_ECMA		European Computer Manufacturers,
73PF_DATAKIT	Datakit protocols,
74PF_CCITT	ITU-T protocols, like X.25,
75PF_SNA		IBM SNA,
76PF_DECnet	DECnet,
77PF_DLI		DEC Direct Data Link Interface protocol,
78PF_LAT		LAT protocol,
79PF_HYLINK	NSC Hyperchannel,
80PF_APPLETALK	AppleTalk protocols,
81PF_ROUTE	Internal Routing protocol,
82PF_LINK		Link layer interface,
83PF_XTP		eXpress Transfer Protocol,
84PF_COIP		Connection-Oriented IP, aka ST II,
85PF_CNT		Computer Network Technology,
86PF_SIP		Simple Internet Protocol,
87PF_IPX		Novell Internet Packet eXchange protocol,
88PF_RTIP		Help Identify RTIP packets,
89PF_PIP		Help Identify PIP packets,
90PF_ISDN		Integrated Services Digital Network,
91PF_KEY		Internal key-management function,
92PF_INET6	Internet version 6 protocols,
93PF_NATM		Native ATM access,
94PF_ATM		ATM,
95PF_NETGRAPH	Netgraph sockets
96.Ed
97.Pp
98The socket has the indicated
99.Fa type ,
100which specifies the semantics of communication.  Currently
101defined types are:
102.Pp
103.Bd -literal -offset indent -compact
104SOCK_STREAM	Stream socket,
105SOCK_DGRAM	Datagram socket,
106SOCK_RAW	Raw-protocol interface,
107SOCK_RDM	Reliably-delivered packet,
108SOCK_SEQPACKET	Sequenced packet stream
109.Ed
110.Pp
111A
112.Dv SOCK_STREAM
113type provides sequenced, reliable,
114two-way connection based byte streams.
115An out-of-band data transmission mechanism may be supported.
116A
117.Dv SOCK_DGRAM
118socket supports
119datagrams (connectionless, unreliable messages of
120a fixed (typically small) maximum length).
121A
122.Dv SOCK_SEQPACKET
123socket may provide a sequenced, reliable,
124two-way connection-based data transmission path for datagrams
125of fixed maximum length; a consumer may be required to read
126an entire packet with each read system call.
127This facility is protocol specific, and presently implemented
128only for
129.Dv PF_NS
130and
131.Dv PF_UNIX .
132.Dv SOCK_RAW
133sockets provide access to internal network protocols and interfaces.
134The types
135.Dv SOCK_RAW ,
136which is available only to the super-user, and
137.Dv SOCK_RDM ,
138which is planned,
139but not yet implemented, are not described here.
140.Pp
141The
142.Fa protocol
143specifies a particular protocol to be used with the socket.
144Normally only a single protocol exists to support a particular
145socket type within a given protocol family.  However, it is possible
146that many protocols may exist, in which case a particular protocol
147must be specified in this manner.  The protocol number to use is
148particular to the
149.Dq "communication domain"
150in which communication
151is to take place; see
152.Xr protocols 5 .
153.Pp
154Sockets of type
155.Dv SOCK_STREAM
156are full-duplex byte streams, similar
157to pipes.  A stream socket must be in a
158.Em connected
159state before any data may be sent or received
160on it.  A connection to another socket is created with a
161.Xr connect 2
162call.
163Once connected, data may be transferred using
164.Xr read 2
165and
166.Xr write 2
167calls or some variant of the
168.Xr send 2
169and
170.Xr recv 2
171calls.
172(Some protocol families, such as the Internet family,
173support the notion of an
174.Dq implied connect ,
175which permits data to be sent piggybacked onto a connect operation by
176using the
177.Xr sendto 2
178call.)
179When a session has been completed a
180.Xr close 2
181may be performed.
182Out-of-band data may also be transmitted as described in
183.Xr send 2
184and received as described in
185.Xr recv 2 .
186.Pp
187The communications protocols used to implement a
188.Dv SOCK_STREAM
189insure that data
190is not lost or duplicated.  If a piece of data for which the
191peer protocol has buffer space cannot be successfully transmitted
192within a reasonable length of time, then
193the connection is considered broken and calls
194will indicate an error with
195-1 returns and with
196.Er ETIMEDOUT
197as the specific code
198in the global variable
199.Va errno .
200The protocols optionally keep sockets
201.Dq warm
202by forcing transmissions
203roughly every minute in the absence of other activity.
204An error is then indicated if no response can be
205elicited on an otherwise
206idle connection for an extended period (e.g. 5 minutes).
207A
208.Dv SIGPIPE
209signal is raised if a process sends
210on a broken stream; this causes naive processes,
211which do not handle the signal, to exit.
212.Pp
213.Dv SOCK_SEQPACKET
214sockets employ the same system calls
215as
216.Dv SOCK_STREAM
217sockets.  The only difference
218is that
219.Xr read 2
220calls will return only the amount of data requested,
221and any remaining in the arriving packet will be discarded.
222.Pp
223.Dv SOCK_DGRAM
224and
225.Dv SOCK_RAW
226sockets allow sending of datagrams to correspondents
227named in
228.Xr send 2
229calls.  Datagrams are generally received with
230.Xr recvfrom 2 ,
231which returns the next datagram with its return address.
232.Pp
233An
234.Xr fcntl 2
235call can be used to specify a process group to receive
236a
237.Dv SIGURG
238signal when the out-of-band data arrives.
239It may also enable non-blocking I/O
240and asynchronous notification of I/O events
241via
242.Dv SIGIO .
243.Pp
244The operation of sockets is controlled by socket level
245.Em options .
246These options are defined in the file
247.In sys/socket.h .
248.Xr Setsockopt 2
249and
250.Xr getsockopt 2
251are used to set and get options, respectively.
252.Sh RETURN VALUES
253Upon successful completion
254.Fn socket
255returns a descriptor referencing the socket.
256Otherwise, -1 is returned and the global variable
257.Va errno
258is set to indicate the error.
259.Sh ERRORS
260The
261.Fn socket
262call fails if:
263.Bl -tag -width Er
264.It Bq Er EPROTONOSUPPORT
265The protocol type or the specified protocol is not supported
266within this domain.
267.It Bq Er EMFILE
268The per-process descriptor table is full.
269.It Bq Er ENFILE
270The system file table is full.
271.It Bq Er EACCES
272Permission to create a socket of the specified type and/or protocol
273is denied.
274.It Bq Er ENOBUFS
275Insufficient buffer space is available.
276The socket cannot be created until sufficient resources are freed.
277.El
278.Sh SEE ALSO
279.Xr accept 2 ,
280.Xr bind 2 ,
281.Xr connect 2 ,
282.Xr getpeername 2 ,
283.Xr getsockname 2 ,
284.Xr getsockopt 2 ,
285.Xr ioctl 2 ,
286.Xr listen 2 ,
287.Xr read 2 ,
288.Xr recv 2 ,
289.Xr select 2 ,
290.Xr send 2 ,
291.Xr shutdown 2 ,
292.Xr socketpair 2 ,
293.Xr write 2 ,
294.Xr getprotoent 3 ,
295.Xr netgraph 4 ,
296.Xr protocols 5
297.Rs
298.%T "An Introductory 4.3 BSD Interprocess Communication Tutorial"
299.%B PS1
300.%N 7
301.Re
302.Rs
303.%T "BSD Interprocess Communication Tutorial"
304.%B PS1
305.%N 8
306.Re
307.Sh HISTORY
308The
309.Fn socket
310function call appeared in
311.Bx 4.2 .
312