xref: /dragonfly/lib/libc/sys/getsockopt.2 (revision b7367ef6)
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.\"     @(#)getsockopt.2	8.4 (Berkeley) 5/2/95
33.\" $FreeBSD: src/lib/libc/sys/getsockopt.2,v 1.12.2.11 2002/01/09 17:44:15 yar Exp $
34.\" $DragonFly: src/lib/libc/sys/getsockopt.2,v 1.5 2007/07/14 21:48:15 swildner Exp $
35.\"
36.Dd May 2, 1995
37.Dt GETSOCKOPT 2
38.Os
39.Sh NAME
40.Nm getsockopt ,
41.Nm setsockopt
42.Nd get and set options on sockets
43.Sh LIBRARY
44.Lb libc
45.Sh SYNOPSIS
46.In sys/types.h
47.In sys/socket.h
48.Ft int
49.Fn getsockopt "int s" "int level" "int optname" "void *optval" "socklen_t *optlen"
50.Ft int
51.Fn setsockopt "int s" "int level" "int optname" "const void *optval" "socklen_t optlen"
52.Sh DESCRIPTION
53.Fn Getsockopt
54and
55.Fn setsockopt
56manipulate the
57.Em options
58associated with a socket.  Options may exist at multiple
59protocol levels; they are always present at the uppermost
60.Dq socket
61level.
62.Pp
63When manipulating socket options the level at which the
64option resides and the name of the option must be specified.
65To manipulate options at the socket level,
66.Fa level
67is specified as
68.Dv SOL_SOCKET .
69To manipulate options at any
70other level the protocol number of the appropriate protocol
71controlling the option is supplied.  For example,
72to indicate that an option is to be interpreted by the
73.Tn TCP
74protocol,
75.Fa level
76should be set to the protocol number of
77.Tn TCP ;
78see
79.Xr getprotoent 3 .
80.Pp
81The parameters
82.Fa optval
83and
84.Fa optlen
85are used to access option values for
86.Fn setsockopt .
87For
88.Fn getsockopt
89they identify a buffer in which the value for the
90requested option(s) are to be returned.  For
91.Fn getsockopt ,
92.Fa optlen
93is a value-result parameter, initially containing the
94size of the buffer pointed to by
95.Fa optval ,
96and modified on return to indicate the actual size of
97the value returned.  If no option value is
98to be supplied or returned,
99.Fa optval
100may be NULL.
101.Pp
102.Fa Optname
103and any specified options are passed uninterpreted to the appropriate
104protocol module for interpretation.
105The include file
106.In sys/socket.h
107contains definitions for
108socket level options, described below.
109Options at other protocol levels vary in format and
110name; consult the appropriate entries in
111section
1124 of the manual.
113.Pp
114Most socket-level options utilize an
115.Fa int
116parameter for
117.Fa optval .
118For
119.Fn setsockopt ,
120the parameter should be non-zero to enable a boolean option,
121or zero if the option is to be disabled.
122.Dv SO_LINGER
123uses a
124.Fa struct linger
125parameter, defined in
126.In sys/socket.h ,
127which specifies the desired state of the option and the
128linger interval (see below).
129.Dv SO_SNDTIMEO
130and
131.Dv SO_RCVTIMEO
132use a
133.Fa struct timeval
134parameter, defined in
135.In sys/time.h .
136.Pp
137The following options are recognized at the socket level.
138Except as noted, each may be examined with
139.Fn getsockopt
140and set with
141.Fn setsockopt .
142.Bl -column SO_ACCEPTFILTER -offset indent
143.It Dv SO_DEBUG Ta "enables recording of debugging information"
144.It Dv SO_REUSEADDR Ta "enables local address reuse"
145.It Dv SO_REUSEPORT Ta "enables duplicate address and port bindings"
146.It Dv SO_KEEPALIVE Ta "enables keep connections alive"
147.It Dv SO_DONTROUTE Ta "enables routing bypass for outgoing messages"
148.It Dv SO_LINGER  Ta "linger on close if data present"
149.It Dv SO_BROADCAST Ta "enables permission to transmit broadcast messages"
150.It Dv SO_OOBINLINE Ta "enables reception of out-of-band data in band"
151.It Dv SO_SNDBUF Ta "set buffer size for output"
152.It Dv SO_RCVBUF Ta "set buffer size for input"
153.It Dv SO_SNDLOWAT Ta "set minimum count for output"
154.It Dv SO_RCVLOWAT Ta "set minimum count for input"
155.It Dv SO_SNDTIMEO Ta "set timeout value for output"
156.It Dv SO_RCVTIMEO Ta "set timeout value for input"
157.It Dv SO_ACCEPTFILTER Ta "set accept filter on listening socket"
158.It Dv SO_TYPE Ta "get the type of the socket (get only)"
159.It Dv SO_ERROR Ta "get and clear error on the socket (get only)"
160.El
161.Pp
162.Dv SO_DEBUG
163enables debugging in the underlying protocol modules.
164.Dv SO_REUSEADDR
165indicates that the rules used in validating addresses supplied
166in a
167.Xr bind 2
168call should allow reuse of local addresses.
169.Dv SO_REUSEPORT
170allows completely duplicate bindings by multiple processes
171if they all set
172.Dv SO_REUSEPORT
173before binding the port.
174This option permits multiple instances of a program to each
175receive UDP/IP multicast or broadcast datagrams destined for the bound port.
176.Dv SO_KEEPALIVE
177enables the
178periodic transmission of messages on a connected socket.  Should the
179connected party fail to respond to these messages, the connection is
180considered broken and processes using the socket are notified via a
181.Dv SIGPIPE
182signal when attempting to send data.
183.Dv SO_DONTROUTE
184indicates that outgoing messages should
185bypass the standard routing facilities.  Instead, messages are directed
186to the appropriate network interface according to the network portion
187of the destination address.
188.Pp
189.Dv SO_LINGER
190controls the action taken when unsent messages
191are queued on socket and a
192.Xr close 2
193is performed.
194If the socket promises reliable delivery of data and
195.Dv SO_LINGER
196is set,
197the system will block the process on the
198.Xr close 2
199attempt until it is able to transmit the data or until it decides it
200is unable to deliver the information (a timeout period, termed the
201linger interval, is specified in seconds in the
202.Fn setsockopt
203call when
204.Dv SO_LINGER
205is requested).
206If
207.Dv SO_LINGER
208is disabled and a
209.Xr close 2
210is issued, the system will process the close in a manner that allows
211the process to continue as quickly as possible.
212.Pp
213The option
214.Dv SO_BROADCAST
215requests permission to send broadcast datagrams
216on the socket.
217Broadcast was a privileged operation in earlier versions of the system.
218With protocols that support out-of-band data, the
219.Dv SO_OOBINLINE
220option
221requests that out-of-band data be placed in the normal data input queue
222as received; it will then be accessible with
223.Xr recv 2
224or
225.Xr read 2
226calls without the
227.Dv MSG_OOB
228flag.
229Some protocols always behave as if this option is set.
230.Dv SO_SNDBUF
231and
232.Dv SO_RCVBUF
233are options to adjust the normal
234buffer sizes allocated for output and input buffers, respectively.
235The buffer size may be increased for high-volume connections,
236or may be decreased to limit the possible backlog of incoming data.
237The system places an absolute maximum on these values, which is accessible
238through the
239.Xr sysctl 3
240MIB variable
241.Va kern.ipc.maxsockbuf .
242.Pp
243.Dv SO_SNDLOWAT
244is an option to set the minimum count for output operations.
245Most output operations process all of the data supplied
246by the call, delivering data to the protocol for transmission
247and blocking as necessary for flow control.
248Nonblocking output operations will process as much data as permitted
249subject to flow control without blocking, but will process no data
250if flow control does not allow the smaller of the low water mark value
251or the entire request to be processed.
252A
253.Xr select 2
254operation testing the ability to write to a socket will return true
255only if the low water mark amount could be processed.
256The default value for
257.Dv SO_SNDLOWAT
258is set to a convenient size for network efficiency, often 1024.
259.Dv SO_RCVLOWAT
260is an option to set the minimum count for input operations.
261In general, receive calls will block until any (non-zero) amount of data
262is received, then return with the smaller of the amount available or the amount
263requested.
264The default value for
265.Dv SO_RCVLOWAT
266is 1.
267If
268.Dv SO_RCVLOWAT
269is set to a larger value, blocking receive calls normally
270wait until they have received the smaller of the low water mark value
271or the requested amount.
272Receive calls may still return less than the low water mark if an error
273occurs, a signal is caught, or the type of data next in the receive queue
274is different from that which was returned.
275.Pp
276.Dv SO_SNDTIMEO
277is an option to set a timeout value for output operations.
278It accepts a
279.Fa struct timeval
280parameter with the number of seconds and microseconds
281used to limit waits for output operations to complete.
282If a send operation has blocked for this much time,
283it returns with a partial count
284or with the error
285.Er EWOULDBLOCK
286if no data were sent.
287In the current implementation, this timer is restarted each time additional
288data are delivered to the protocol,
289implying that the limit applies to output portions ranging in size
290from the low water mark to the high water mark for output.
291.Dv SO_RCVTIMEO
292is an option to set a timeout value for input operations.
293It accepts a
294.Fa struct timeval
295parameter with the number of seconds and microseconds
296used to limit waits for input operations to complete.
297In the current implementation, this timer is restarted each time additional
298data are received by the protocol,
299and thus the limit is in effect an inactivity timer.
300If a receive operation has been blocked for this much time without
301receiving additional data, it returns with a short count
302or with the error
303.Er EWOULDBLOCK
304if no data were received.
305.Pp
306.Dv SO_ACCEPTFILTER
307places an
308.Xr accept_filter 9
309on the socket,
310which will filter incoming connections
311on a listening stream socket before being presented for
312.Xr accept 2 .
313Once more,
314.Xr listen 2
315must be called on the socket before
316trying to install the filter on it,
317or else the
318.Fn setsockopt
319call will fail.
320.Bd -literal
321struct  accept_filter_arg {
322        char    af_name[16];
323        char    af_arg[256-16];
324};
325.Ed
326.Pp
327.Fa optval
328should point to a
329.Fa struct accept_filter_arg
330that will select and configure the
331.Xr accept_filter 9 .
332.Fa af_name
333should be filled with the name of the accept filter
334that the application wishes to place on the listening socket.
335.Fa af_arg
336is an optional parameter that can be passed to the accept
337filter specified by
338.Fa af_name
339to provide additional configuration options at attach time.
340Passing in an
341.Fa optval
342of NULL will remove the filter.
343.Pp
344Finally,
345.Dv SO_TYPE
346and
347.Dv SO_ERROR
348are options used only with
349.Fn getsockopt .
350.Dv SO_TYPE
351returns the type of the socket, such as
352.Dv SOCK_STREAM ;
353it is useful for servers that inherit sockets on startup.
354.Dv SO_ERROR
355returns any pending error on the socket and clears
356the error status.
357It may be used to check for asynchronous errors on connected
358datagram sockets or for other asynchronous errors.
359.Sh RETURN VALUES
360.Rv -std
361.Sh ERRORS
362The call succeeds unless:
363.Bl -tag -width Er
364.It Bq Er EBADF
365The argument
366.Fa s
367is not a valid descriptor.
368.It Bq Er ENOTSOCK
369The argument
370.Fa s
371is a file, not a socket.
372.It Bq Er ENOPROTOOPT
373The option is unknown at the level indicated.
374.It Bq Er EFAULT
375The address pointed to by
376.Fa optval
377is not in a valid part of the process address space.
378For
379.Fn getsockopt ,
380this error may also be returned if
381.Fa optlen
382is not in a valid part of the process address space.
383.It Bq Er EINVAL
384Installing an
385.Xr accept_filter 9
386on a non-listening socket was attempted.
387.El
388.Sh SEE ALSO
389.Xr ioctl 2 ,
390.Xr socket 2 ,
391.Xr getprotoent 3 ,
392.Xr sysctl 3 ,
393.Xr protocols 5 ,
394.Xr sysctl 8 ,
395.Xr accept_filter 9
396.Sh HISTORY
397The
398.Fn getsockopt
399system call appeared in
400.Bx 4.2 .
401.Sh BUGS
402Several of the socket options should be handled at lower levels of the system.
403