1SOCKET(7) Linux Programmer's Manual SOCKET(7)
2
6 socket - Linux socket interface
7
9 #include <sys/socket.h>
10 sockfd = socket(int socket_family, int socket_type, int protocol);
12
14 This manual page describes the Linux networking socket layer user
15 interface. The BSD compatible sockets are the uniform interface
16 between the user process and the network protocol stacks in the kernel.
17 The protocol modules are grouped into protocol families like AF_INET,
18 AF_IPX, AF_PACKET and socket types like SOCK_STREAM or SOCK_DGRAM. See
19 socket(2) for more information on families and types.
20 Socket Layer Functions
22 These functions are used by the user process to send or receive packets
23 and to do other socket operations. For more information see their
24 respective manual pages.
25 socket(2) creates a socket, connect(2) connects a socket to a remote
27 socket address, the bind(2) function binds a socket to a local socket
28 address, listen(2) tells the socket that new connections shall be
29 accepted, and accept(2) is used to get a new socket with a new incoming
30 connection. socketpair(2) returns two connected anonymous sockets
31 (only implemented for a few local families like AF_UNIX)
32 send(2), sendto(2), and sendmsg(2) send data over a socket, and
34 recv(2), recvfrom(2), recvmsg(2) receive data from a socket. poll(2)
35 and select(2) wait for arriving data or a readiness to send data. In
36 addition, the standard I/O operations like write(2), writev(2), send‐
37 file(2), read(2), and readv(2) can be used to read and write data.
38 getsockname(2) returns the local socket address and getpeername(2)
40 returns the remote socket address. getsockopt(2) and setsockopt(2) are
41 used to set or get socket layer or protocol options. ioctl(2) can be
42 used to set or read some other options.
43 close(2) is used to close a socket. shutdown(2) closes parts of a
45 full-duplex socket connection.
46 Seeking, or calling pread(2) or pwrite(2) with a non-zero position is
48 not supported on sockets.
49 It is possible to do non-blocking I/O on sockets by setting the O_NON‐
51 BLOCK flag on a socket file descriptor using fcntl(2). Then all opera‐
52 tions that would block will (usually) return with EAGAIN (operation
53 should be retried later); connect(2) will return EINPROGRESS error.
54 The user can then wait for various events via poll(2) or select(2).
55 ┌────────────────────────────────────────────────────────────────────┐
57 │ I/O events │
58 ├───────────┬───────────┬────────────────────────────────────────────┤
59 │Event │ Poll flag │ Occurrence │
60 ├───────────┼───────────┼────────────────────────────────────────────┤
61 │Read │ POLLIN │ New data arrived. │
62 ├───────────┼───────────┼────────────────────────────────────────────┤
63 │Read │ POLLIN │ A connection setup has been completed (for │
64 │ │ │ connection-oriented sockets) │
65 ├───────────┼───────────┼────────────────────────────────────────────┤
66 │Read │ POLLHUP │ A disconnection request has been initiated │
67 │ │ │ by the other end. │
68 ├───────────┼───────────┼────────────────────────────────────────────┤
69 │Read │ POLLHUP │ A connection is broken (only for connec‐ │
70 │ │ │ tion-oriented protocols). When the socket │
71 │ │ │ is written SIGPIPE is also sent. │
72 ├───────────┼───────────┼────────────────────────────────────────────┤
73 │Write │ POLLOUT │ Socket has enough send buffer space for │
74 │ │ │ writing new data. │
75 ├───────────┼───────────┼────────────────────────────────────────────┤
76 │Read/Write │ POLLIN| │ An outgoing connect(2) finished. │
77 │ │ POLLOUT │ │
78 ├───────────┼───────────┼────────────────────────────────────────────┤
79 │Read/Write │ POLLERR │ An asynchronous error occurred. │
80 ├───────────┼───────────┼────────────────────────────────────────────┤
81 │Read/Write │ POLLHUP │ The other end has shut down one direction. │
82 ├───────────┼───────────┼────────────────────────────────────────────┤
83 │Exception │ POLLPRI │ Urgent data arrived. SIGURG is sent then. │
84 └───────────┴───────────┴────────────────────────────────────────────┘
85 An alternative to poll(2) and select(2) is to let the kernel inform the
87 application about events via a SIGIO signal. For that the O_ASYNC flag
88 must be set on a socket file descriptor via fcntl(2) and a valid signal
89 handler for SIGIO must be installed via sigaction(2). See the Signals
90 discussion below.
91 Socket Options
93 These socket options can be set by using setsockopt(2) and read with
94 getsockopt(2) with the socket level set to SOL_SOCKET for all sockets:
95 SO_ACCEPTCONN
97 Returns a value indicating whether or not this socket has been
98 marked to accept connections with listen(2). The value 0 indi‐
99 cates that this is not a listening socket, the value 1 indicates
100 that this is a listening socket. Can only be read with getsock‐
101 opt(2).
102 SO_BINDTODEVICE
104 Bind this socket to a particular device like “eth0”, as speci‐
105 fied in the passed interface name. If the name is an empty
106 string or the option length is zero, the socket device binding
107 is removed. The passed option is a variable-length null-termi‐
108 nated interface name string with the maximum size of IFNAMSIZ.
109 If a socket is bound to an interface, only packets received from
110 that particular interface are processed by the socket. Note
111 that this only works for some socket types, particularly AF_INET
112 sockets. It is not supported for packet sockets (use normal
113 bind(8) there).
114 SO_BROADCAST
116 Set or get the broadcast flag. When enabled, datagram sockets
117 receive packets sent to a broadcast address and they are allowed
118 to send packets to a broadcast address. This option has no
119 effect on stream-oriented sockets.
120 SO_BSDCOMPAT
122 Enable BSD bug-to-bug compatibility. This is used by the UDP
123 protocol module in Linux 2.0 and 2.2. If enabled ICMP errors
124 received for a UDP socket will not be passed to the user pro‐
125 gram. In later kernel versions, support for this option has
126 been phased out: Linux 2.4 silently ignores it, and Linux 2.6
127 generates a kernel warning (printk()) if a program uses this
128 option. Linux 2.0 also enabled BSD bug-to-bug compatibility
129 options (random header changing, skipping of the broadcast flag)
130 for raw sockets with this option, but that was removed in Linux
131 2.2.
132 SO_DEBUG
134 Enable socket debugging. Only allowed for processes with the
135 CAP_NET_ADMIN capability or an effective user ID of 0.
136 SO_ERROR
138 Get and clear the pending socket error. Only valid as a get‐
139 sockopt(2). Expects an integer.
140 SO_DONTROUTE
142 Don't send via a gateway, only send to directly connected hosts.
143 The same effect can be achieved by setting the MSG_DONTROUTE
144 flag on a socket send(2) operation. Expects an integer boolean
145 flag.
146 SO_KEEPALIVE
148 Enable sending of keep-alive messages on connection-oriented
149 sockets. Expects an integer boolean flag.
150 SO_LINGER
152 Sets or gets the SO_LINGER option. The argument is a linger
153 structure.
154 struct linger {
156 int l_onoff; /* linger active */
157 int l_linger; /* how many seconds to linger for */
158 };
159 When enabled, a close(2) or shutdown(2) will not return until
161 all queued messages for the socket have been successfully sent
162 or the linger timeout has been reached. Otherwise, the call
163 returns immediately and the closing is done in the background.
164 When the socket is closed as part of exit(2), it always lingers
165 in the background.
166 SO_OOBINLINE
168 If this option is enabled, out-of-band data is directly placed
169 into the receive data stream. Otherwise out-of-band data is
170 only passed when the MSG_OOB flag is set during receiving.
171 SO_PASSCRED
173 Enable or disable the receiving of the SCM_CREDENTIALS control
174 message. For more information see unix(7).
175 SO_PEERCRED
177 Return the credentials of the foreign process connected to this
178 socket. This is only possible for connected AF_UNIX stream
179 sockets and AF_UNIX stream and datagram socket pairs created
180 using socketpair(2); see unix(7). The returned credentials are
181 those that were in effect at the time of the call to connect(2)
182 or socketpair(2). Argument is a ucred structure. Only valid as
183 a getsockopt(2).
184 SO_PRIORITY
186 Set the protocol-defined priority for all packets to be sent on
187 this socket. Linux uses this value to order the networking
188 queues: packets with a higher priority may be processed first
189 depending on the selected device queueing discipline. For
190 ip(7), this also sets the IP type-of-service (TOS) field for
191 outgoing packets. Setting a priority outside the range 0 to 6
192 requires the CAP_NET_ADMIN capability.
193 SO_RCVBUF
195 Sets or gets the maximum socket receive buffer in bytes. The
196 kernel doubles this value (to allow space for bookkeeping over‐
197 head) when it is set using setsockopt(2), and this doubled value
198 is returned by getsockopt(2). The default value is set by the
199 /proc/sys/net/core/rmem_default file, and the maximum allowed
200 value is set by the /proc/sys/net/core/rmem_max file. The mini‐
201 mum (doubled) value for this option is 256.
202 SO_RCVBUFFORCE (since Linux 2.6.14)
204 Using this socket option, a privileged (CAP_NET_ADMIN) process
205 can perform the same task as SO_RCVBUF, but the rmem_max limit
206 can be overridden.
207 SO_RCVLOWAT and SO_SNDLOWAT
209 Specify the minimum number of bytes in the buffer until the
210 socket layer will pass the data to the protocol (SO_SNDLOWAT) or
211 the user on receiving (SO_RCVLOWAT). These two values are ini‐
212 tialized to 1. SO_SNDLOWAT is not changeable on Linux (setsock‐
213 opt(2) fails with the error ENOPROTOOPT). SO_RCVLOWAT is
214 changeable only since Linux 2.4. The select(2) and poll(2) sys‐
215 tem calls currently do not respect the SO_RCVLOWAT setting on
216 Linux, and mark a socket readable when even a single byte of
217 data is available. A subsequent read from the socket will block
218 until SO_RCVLOWAT bytes are available.
219 SO_RCVTIMEO and SO_SNDTIMEO
221 Specify the receiving or sending timeouts until reporting an
222 error. The argument is a struct timeval. If an input or output
223 function blocks for this period of time, and data has been sent
224 or received, the return value of that function will be the
225 amount of data transferred; if no data has been transferred and
226 the timeout has been reached then -1 is returned with errno set
227 to EAGAIN or EWOULDBLOCK just as if the socket was specified to
228 be non-blocking. If the timeout is set to zero (the default)
229 then the operation will never timeout. Timeouts only have
230 effect for system calls that perform socket I/O (e.g., read(2),
231 recvmsg(2), send(2), sendmsg(2)); timeouts have no effect for
232 select(2), poll(2), epoll_wait(2), etc.
233 SO_REUSEADDR
235 Indicates that the rules used in validating addresses supplied
236 in a bind(2) call should allow reuse of local addresses. For
237 AF_INET sockets this means that a socket may bind, except when
238 there is an active listening socket bound to the address. When
239 the listening socket is bound to INADDR_ANY with a specific port
240 then it is not possible to bind to this port for any local
241 address. Argument is an integer boolean flag.
242 SO_SNDBUF
244 Sets or gets the maximum socket send buffer in bytes. The ker‐
245 nel doubles this value (to allow space for bookkeeping overhead)
246 when it is set using setsockopt(2), and this doubled value is
247 returned by getsockopt(2). The default value is set by the
248 /proc/sys/net/core/wmem_default file and the maximum allowed
249 value is set by the /proc/sys/net/core/wmem_max file. The mini‐
250 mum (doubled) value for this option is 2048.
251 SO_SNDBUFFORCE (since Linux 2.6.14)
253 Using this socket option, a privileged (CAP_NET_ADMIN) process
254 can perform the same task as SO_SNDBUF, but the wmem_max limit
255 can be overridden.
256 SO_TIMESTAMP
258 Enable or disable the receiving of the SO_TIMESTAMP control mes‐
259 sage. The timestamp control message is sent with level
260 SOL_SOCKET and the cmsg_data field is a struct timeval indicat‐
261 ing the reception time of the last packet passed to the user in
262 this call. See cmsg(3) for details on control messages.
263 SO_TYPE
265 Gets the socket type as an integer (like SOCK_STREAM). Can only
266 be read with getsockopt(2).
267 Signals
269 When writing onto a connection-oriented socket that has been shut down
270 (by the local or the remote end) SIGPIPE is sent to the writing process
271 and EPIPE is returned. The signal is not sent when the write call
272 specified the MSG_NOSIGNAL flag.
273 When requested with the FIOSETOWN fcntl(2) or SIOCSPGRP ioctl(2), SIGIO
275 is sent when an I/O event occurs. It is possible to use poll(2) or
276 select(2) in the signal handler to find out which socket the event
277 occurred on. An alternative (in Linux 2.2) is to set a real-time sig‐
278 nal using the F_SETSIG fcntl(2); the handler of the real time signal
279 will be called with the file descriptor in the si_fd field of its sig‐
280 info_t. See fcntl(2) for more information.
281 Under some circumstances (e.g., multiple processes accessing a single
283 socket), the condition that caused the SIGIO may have already disap‐
284 peared when the process reacts to the signal. If this happens, the
285 process should wait again because Linux will resend the signal later.
286 /proc interfaces
288 The core socket networking parameters can be accessed via files in the
289 directory /proc/sys/net/core/.
290 rmem_default
292 contains the default setting in bytes of the socket receive buf‐
293 fer.
294 rmem_max
296 contains the maximum socket receive buffer size in bytes which a
297 user may set by using the SO_RCVBUF socket option.
298 wmem_default
300 contains the default setting in bytes of the socket send buffer.
301 wmem_max
303 contains the maximum socket send buffer size in bytes which a
304 user may set by using the SO_SNDBUF socket option.
305 message_cost and message_burst
307 configure the token bucket filter used to load limit warning
308 messages caused by external network events.
309 netdev_max_backlog
311 Maximum number of packets in the global input queue.
312 optmem_max
314 Maximum length of ancillary data and user control data like the
315 iovecs per socket.
316 Ioctls
318 These operations can be accessed using ioctl(2):
319 error = ioctl(ip_socket, ioctl_type, &value_result);
321 SIOCGSTAMP
323 Return a struct timeval with the receive timestamp of the last
324 packet passed to the user. This is useful for accurate round
325 trip time measurements. See setitimer(2) for a description of
326 struct timeval. This ioctl should only be used if the socket
327 option SO_TIMESTAMP is not set on the socket. Otherwise, it
328 returns the timestamp of the last packet that was received while
329 SO_TIMESTAMP was not set, or it fails if no such packet has been
330 received, (i.e., ioctl(2) returns -1 with errno set to ENOENT).
331 SIOCSPGRP
333 Set the process or process group to send SIGIO or SIGURG signals
334 to when an asynchronous I/O operation has finished or urgent
335 data is available. The argument is a pointer to a pid_t. If
336 the argument is positive, send the signals to that process. If
337 the argument is negative, send the signals to the process group
338 with the ID of the absolute value of the argument. The process
339 may only choose itself or its own process group to receive sig‐
340 nals unless it has the CAP_KILL capability or an effective UID
341 of 0.
342 FIOASYNC
344 Change the O_ASYNC flag to enable or disable asynchronous I/O
345 mode of the socket. Asynchronous I/O mode means that the SIGIO
346 signal or the signal set with F_SETSIG is raised when a new I/O
347 event occurs.
348 Argument is an integer boolean flag. (This operation is synony‐
350 mous with the use of fcntl(2) to set the O_ASYNC flag.)
351 SIOCGPGRP
353 Get the current process or process group that receives SIGIO or
354 SIGURG signals, or 0 when none is set.
355 Valid fcntl(2) operations:
357 FIOGETOWN
359 The same as the SIOCGPGRP ioctl(2).
360 FIOSETOWN
362 The same as the SIOCSPGRP ioctl(2).
363
365 SO_BINDTODEVICE was introduced in Linux 2.0.30. SO_PASSCRED is new in
366 Linux 2.2. The /proc interfaces was introduced in Linux 2.2. SO_RCV‐
367 TIMEO and SO_SNDTIMEO are supported since Linux 2.3.41. Earlier, time‐
368 outs were fixed to a protocol-specific setting, and could not be read
369 or written.
370
372 Linux assumes that half of the send/receive buffer is used for internal
373 kernel structures; thus the values in the corresponding /proc files are
374 twice what can be observed on the wire.
375 Linux will only allow port re-use with the SO_REUSEADDR option when
377 this option was set both in the previous program that performed a
378 bind(2) to the port and in the program that wants to re-use the port.
379 This differs from some implementations (e.g., FreeBSD) where only the
380 later program needs to set the SO_REUSEADDR option. Typically this
381 difference is invisible, since, for example, a server program is
382 designed to always set this option.
383
385 The CONFIG_FILTER socket options SO_ATTACH_FILTER and SO_DETACH_FILTER
386 are not documented. The suggested interface to use them is via the
387 libpcap library.
388
390 getsockopt(2), setsockopt(2), socket(2), capabilities(7), ddp(7),
391 ip(7), packet(7), tcp(7), udp(7), unix(7)
392
394 This page is part of release 3.22 of the Linux man-pages project. A
395 description of the project, and information about reporting bugs, can
396 be found at http://www.kernel.org/doc/man-pages/.
397Linux 2008-12-03 SOCKET(7)