1SEND(2) Linux Programmer's Manual SEND(2)
2
6 send, sendto, sendmsg - send a message on a socket
7
9 #include <sys/types.h>
10 #include <sys/socket.h>
11 ssize_t send(int sockfd, const void *buf, size_t len, int flags);
13 ssize_t sendto(int sockfd, const void *buf, size_t len, int flags,
15 const struct sockaddr *dest_addr, socklen_t addrlen);
16 ssize_t sendmsg(int sockfd, const struct msghdr *msg, int flags);
18
20 The system calls send(), sendto(), and sendmsg() are used to transmit a
21 message to another socket.
22 The send() call may be used only when the socket is in a connected
24 state (so that the intended recipient is known). The only difference
25 between send() and write(2) is the presence of flags. With a zero
26 flags argument, send() is equivalent to write(2). Also, the following
27 call
28 send(sockfd, buf, len, flags);
30 is equivalent to
32 sendto(sockfd, buf, len, flags, NULL, 0);
34 The argument sockfd is the file descriptor of the sending socket.
36 If sendto() is used on a connection-mode (SOCK_STREAM, SOCK_SEQPACKET)
38 socket, the arguments dest_addr and addrlen are ignored (and the error
39 EISCONN may be returned when they are not NULL and 0), and the error
40 ENOTCONN is returned when the socket was not actually connected. Oth‐
41 erwise, the address of the target is given by dest_addr with addrlen
42 specifying its size. For sendmsg(), the address of the target is given
43 by msg.msg_name, with msg.msg_namelen specifying its size.
44 For send() and sendto(), the message is found in buf and has length
46 len. For sendmsg(), the message is pointed to by the elements of the
47 array msg.msg_iov. The sendmsg() call also allows sending ancillary
48 data (also known as control information).
49 If the message is too long to pass atomically through the underlying
51 protocol, the error EMSGSIZE is returned, and the message is not trans‐
52 mitted.
53 No indication of failure to deliver is implicit in a send(). Locally
55 detected errors are indicated by a return value of -1.
56 When the message does not fit into the send buffer of the socket,
58 send() normally blocks, unless the socket has been placed in nonblock‐
59 ing I/O mode. In nonblocking mode it would fail with the error EAGAIN
60 or EWOULDBLOCK in this case. The select(2) call may be used to deter‐
61 mine when it is possible to send more data.
62 The flags argument
64 The flags argument is the bitwise OR of zero or more of the following
65 flags.
66 MSG_CONFIRM (since Linux 2.3.15)
68 Tell the link layer that forward progress happened: you got a
69 successful reply from the other side. If the link layer doesn't
70 get this it will regularly reprobe the neighbor (e.g., via a
71 unicast ARP). Valid only on SOCK_DGRAM and SOCK_RAW sockets and
72 currently implemented only for IPv4 and IPv6. See arp(7) for
73 details.
74 MSG_DONTROUTE
76 Don't use a gateway to send out the packet, send to hosts only
77 on directly connected networks. This is usually used only by
78 diagnostic or routing programs. This is defined only for proto‐
79 col families that route; packet sockets don't.
80 MSG_DONTWAIT (since Linux 2.2)
82 Enables nonblocking operation; if the operation would block,
83 EAGAIN or EWOULDBLOCK is returned. This provides similar behav‐
84 ior to setting the O_NONBLOCK flag (via the fcntl(2) F_SETFL
85 operation), but differs in that MSG_DONTWAIT is a per-call
86 option, whereas O_NONBLOCK is a setting on the open file
87 description (see open(2)), which will affect all threads in the
88 calling process and as well as other processes that hold file
89 descriptors referring to the same open file description.
90 MSG_EOR (since Linux 2.2)
92 Terminates a record (when this notion is supported, as for sock‐
93 ets of type SOCK_SEQPACKET).
94 MSG_MORE (since Linux 2.4.4)
96 The caller has more data to send. This flag is used with TCP
97 sockets to obtain the same effect as the TCP_CORK socket option
98 (see tcp(7)), with the difference that this flag can be set on a
99 per-call basis.
100 Since Linux 2.6, this flag is also supported for UDP sockets,
102 and informs the kernel to package all of the data sent in calls
103 with this flag set into a single datagram which is transmitted
104 only when a call is performed that does not specify this flag.
105 (See also the UDP_CORK socket option described in udp(7).)
106 MSG_NOSIGNAL (since Linux 2.2)
108 Don't generate a SIGPIPE signal if the peer on a stream-oriented
109 socket has closed the connection. The EPIPE error is still
110 returned. This provides similar behavior to using sigaction(2)
111 to ignore SIGPIPE, but, whereas MSG_NOSIGNAL is a per-call fea‐
112 ture, ignoring SIGPIPE sets a process attribute that affects all
113 threads in the process.
114 MSG_OOB
116 Sends out-of-band data on sockets that support this notion
117 (e.g., of type SOCK_STREAM); the underlying protocol must also
118 support out-of-band data.
119 sendmsg()
121 The definition of the msghdr structure employed by sendmsg() is as fol‐
122 lows:
123 struct msghdr {
125 void *msg_name; /* Optional address */
126 socklen_t msg_namelen; /* Size of address */
127 struct iovec *msg_iov; /* Scatter/gather array */
128 size_t msg_iovlen; /* # elements in msg_iov */
129 void *msg_control; /* Ancillary data, see below */
130 size_t msg_controllen; /* Ancillary data buffer len */
131 int msg_flags; /* Flags (unused) */
132 };
133 The msg_name field is used on an unconnected socket to specify the tar‐
135 get address for a datagram. It points to a buffer containing the
136 address; the msg_namelen field should be set to the size of the
137 address. For a connected socket, these fields should be specified as
138 NULL and 0, respectively.
139 The msg_iov and msg_iovlen fields specify scatter-gather locations, as
141 for writev(2).
142 You may send control information using the msg_control and msg_con‐
144 trollen members. The maximum control buffer length the kernel can
145 process is limited per socket by the value in /proc/sys/net/core/opt‐
146 mem_max; see socket(7).
147 The msg_flags field is ignored.
149
151 On success, these calls return the number of bytes sent. On error, -1
152 is returned, and errno is set appropriately.
153
155 These are some standard errors generated by the socket layer. Addi‐
156 tional errors may be generated and returned from the underlying proto‐
157 col modules; see their respective manual pages.
158 EACCES (For UNIX domain sockets, which are identified by pathname)
160 Write permission is denied on the destination socket file, or
161 search permission is denied for one of the directories the path
162 prefix. (See path_resolution(7).)
163 (For UDP sockets) An attempt was made to send to a net‐
165 work/broadcast address as though it was a unicast address.
166 EAGAIN or EWOULDBLOCK
168 The socket is marked nonblocking and the requested operation
169 would block. POSIX.1-2001 allows either error to be returned
170 for this case, and does not require these constants to have the
171 same value, so a portable application should check for both pos‐
172 sibilities.
173 EAGAIN (Internet domain datagram sockets) The socket referred to by
175 sockfd had not previously been bound to an address and, upon
176 attempting to bind it to an ephemeral port, it was determined
177 that all port numbers in the ephemeral port range are currently
178 in use. See the discussion of
179 /proc/sys/net/ipv4/ip_local_port_range in ip(7).
180 EALREADY
182 Another Fast Open is in progress.
183 EBADF sockfd is not a valid open file descriptor.
185 ECONNRESET
187 Connection reset by peer.
188 EDESTADDRREQ
190 The socket is not connection-mode, and no peer address is set.
191 EFAULT An invalid user space address was specified for an argument.
193 EINTR A signal occurred before any data was transmitted; see sig‐
195 nal(7).
196 EINVAL Invalid argument passed.
198 EISCONN
200 The connection-mode socket was connected already but a recipient
201 was specified. (Now either this error is returned, or the
202 recipient specification is ignored.)
203 EMSGSIZE
205 The socket type requires that message be sent atomically, and
206 the size of the message to be sent made this impossible.
207 ENOBUFS
209 The output queue for a network interface was full. This gener‐
210 ally indicates that the interface has stopped sending, but may
211 be caused by transient congestion. (Normally, this does not
212 occur in Linux. Packets are just silently dropped when a device
213 queue overflows.)
214 ENOMEM No memory available.
216 ENOTCONN
218 The socket is not connected, and no target has been given.
219 ENOTSOCK
221 The file descriptor sockfd does not refer to a socket.
222 EOPNOTSUPP
224 Some bit in the flags argument is inappropriate for the socket
225 type.
226 EPIPE The local end has been shut down on a connection oriented
228 socket. In this case, the process will also receive a SIGPIPE
229 unless MSG_NOSIGNAL is set.
230
232 4.4BSD, SVr4, POSIX.1-2001. These interfaces first appeared in 4.2BSD.
233 POSIX.1-2001 describes only the MSG_OOB and MSG_EOR flags.
235 POSIX.1-2008 adds a specification of MSG_NOSIGNAL. The MSG_CONFIRM
236 flag is a Linux extension.
237
239 According to POSIX.1-2001, the msg_controllen field of the msghdr
240 structure should be typed as socklen_t, but glibc currently types it as
241 size_t.
242 See sendmmsg(2) for information about a Linux-specific system call that
244 can be used to transmit multiple datagrams in a single call.
245
247 Linux may return EPIPE instead of ENOTCONN.
248
250 An example of the use of sendto() is shown in getaddrinfo(3).
251
253 fcntl(2), getsockopt(2), recv(2), select(2), sendfile(2), sendmmsg(2),
254 shutdown(2), socket(2), write(2), cmsg(3), ip(7), ipv6(7), socket(7),
255 tcp(7), udp(7), unix(7)
256
258 This page is part of release 5.02 of the Linux man-pages project. A
259 description of the project, information about reporting bugs, and the
260 latest version of this page, can be found at
261 https://www.kernel.org/doc/man-pages/.
262Linux 2017-09-15 SEND(2)