diff options
Diffstat (limited to '')
| -rw-r--r-- | man2/send.2 | 506 |
1 files changed, 506 insertions, 0 deletions
diff --git a/man2/send.2 b/man2/send.2 new file mode 100644 index 0000000..16c58b5 --- /dev/null +++ b/man2/send.2 @@ -0,0 +1,506 @@ +.\" Copyright (c) 1983, 1991 The Regents of the University of California. +.\" All rights reserved. +.\" +.\" SPDX-License-Identifier: BSD-4-Clause-UC +.\" +.\" Modified 1993-07-24 by Rik Faith <faith@cs.unc.edu> +.\" Modified 1996-10-22 by Eric S. Raymond <esr@thyrsus.com> +.\" Modified Oct 1998 by Andi Kleen +.\" Modified Oct 2003 by aeb +.\" Modified 2004-07-01 by mtk +.\" +.TH send 2 2023-03-30 "Linux man-pages 6.05.01" +.SH NAME +send, sendto, sendmsg \- send a message on a socket +.SH LIBRARY +Standard C library +.RI ( libc ", " \-lc ) +.SH SYNOPSIS +.nf +.B #include <sys/socket.h> +.PP +.BI "ssize_t send(int " sockfd ", const void " buf [. len "], size_t " len \ +", int " flags ); +.BI "ssize_t sendto(int " sockfd ", const void " buf [. len "], size_t " len \ +", int " flags , +.BI " const struct sockaddr *" dest_addr ", socklen_t " addrlen ); +.BI "ssize_t sendmsg(int " sockfd ", const struct msghdr *" msg \ +", int " flags ); +.fi +.SH DESCRIPTION +The system calls +.BR send (), +.BR sendto (), +and +.BR sendmsg () +are used to transmit a message to another socket. +.PP +The +.BR send () +call may be used only when the socket is in a +.I connected +state (so that the intended recipient is known). +The only difference between +.BR send () +and +.BR write (2) +is the presence of +.IR flags . +With a zero +.I flags +argument, +.BR send () +is equivalent to +.BR write (2). +Also, the following call +.PP +.in +4n +.EX +send(sockfd, buf, len, flags); +.EE +.in +.PP +is equivalent to +.PP +.in +4n +.EX +sendto(sockfd, buf, len, flags, NULL, 0); +.EE +.in +.PP +The argument +.I sockfd +is the file descriptor of the sending socket. +.PP +If +.BR sendto () +is used on a connection-mode +.RB ( SOCK_STREAM , +.BR SOCK_SEQPACKET ) +socket, the arguments +.I dest_addr +and +.I addrlen +are ignored (and the error +.B EISCONN +may be returned when they are +not NULL and 0), and the error +.B ENOTCONN +is returned when the socket was not actually connected. +Otherwise, the address of the target is given by +.I dest_addr +with +.I addrlen +specifying its size. +For +.BR sendmsg (), +the address of the target is given by +.IR msg.msg_name , +with +.I msg.msg_namelen +specifying its size. +.PP +For +.BR send () +and +.BR sendto (), +the message is found in +.I buf +and has length +.IR len . +For +.BR sendmsg (), +the message is pointed to by the elements of the array +.IR msg.msg_iov . +The +.BR sendmsg () +call also allows sending ancillary data (also known as control information). +.PP +If the message is too long to pass atomically through the +underlying protocol, the error +.B EMSGSIZE +is returned, and the message is not transmitted. +.PP +No indication of failure to deliver is implicit in a +.BR send (). +Locally detected errors are indicated by a return value of \-1. +.PP +When the message does not fit into the send buffer of the socket, +.BR send () +normally blocks, unless the socket has been placed in nonblocking I/O +mode. +In nonblocking mode it would fail with the error +.B EAGAIN +or +.B EWOULDBLOCK +in this case. +The +.BR select (2) +call may be used to determine when it is possible to send more data. +.SS The flags argument +The +.I flags +argument is the bitwise OR +of zero or more of the following flags. +.\" FIXME . ? document MSG_PROXY (which went away in Linux 2.3.15) +.TP +.BR MSG_CONFIRM " (since Linux 2.3.15)" +Tell the link layer that forward progress happened: you got a successful +reply from the other side. +If the link layer doesn't get this +it will regularly reprobe the neighbor (e.g., via a unicast ARP). +Valid only on +.B SOCK_DGRAM +and +.B SOCK_RAW +sockets and currently implemented only for IPv4 and IPv6. +See +.BR arp (7) +for details. +.TP +.B MSG_DONTROUTE +Don't use a gateway to send out the packet, send to hosts only on +directly connected networks. +This is usually used only +by diagnostic or routing programs. +This is defined only for protocol +families that route; packet sockets don't. +.TP +.BR MSG_DONTWAIT " (since Linux 2.2)" +Enables nonblocking operation; if the operation would block, +.B EAGAIN +or +.B EWOULDBLOCK +is returned. +This provides similar behavior to setting the +.B O_NONBLOCK +flag (via the +.BR fcntl (2) +.B F_SETFL +operation), but differs in that +.B MSG_DONTWAIT +is a per-call option, whereas +.B O_NONBLOCK +is a setting on the open file description (see +.BR open (2)), +which will affect all threads in the calling process +and as well as other processes that hold file descriptors +referring to the same open file description. +.TP +.BR MSG_EOR " (since Linux 2.2)" +Terminates a record (when this notion is supported, as for sockets of type +.BR SOCK_SEQPACKET ). +.TP +.BR MSG_MORE " (since Linux 2.4.4)" +The caller has more data to send. +This flag is used with TCP sockets to obtain the same effect +as the +.B TCP_CORK +socket option (see +.BR tcp (7)), +with the difference that this flag can be set on a per-call basis. +.IP +Since Linux 2.6, this flag is also supported for UDP sockets, and informs +the kernel to package all of the data sent in calls with this flag set +into a single datagram which is transmitted only when a call is performed +that does not specify this flag. +(See also the +.B UDP_CORK +socket option described in +.BR udp (7).) +.TP +.BR MSG_NOSIGNAL " (since Linux 2.2)" +Don't generate a +.B SIGPIPE +signal if the peer on a stream-oriented socket has closed the connection. +The +.B EPIPE +error is still returned. +This provides similar behavior to using +.BR sigaction (2) +to ignore +.BR SIGPIPE , +but, whereas +.B MSG_NOSIGNAL +is a per-call feature, +ignoring +.B SIGPIPE +sets a process attribute that affects all threads in the process. +.TP +.B MSG_OOB +Sends +.I out-of-band +data on sockets that support this notion (e.g., of type +.BR SOCK_STREAM ); +the underlying protocol must also support +.I out-of-band +data. +.TP +.BR MSG_FASTOPEN " (since Linux 3.7)" +Attempts TCP Fast Open (RFC7413) and sends data in the SYN like a +combination of +.BR connect (2) +and +.BR write (2), +by performing an implicit +.BR connect (2) +operation. +It blocks until the data is buffered and the handshake has completed. +For a non-blocking socket, +it returns the number of bytes buffered and sent in the SYN packet. +If the cookie is not available locally, +it returns +.BR EINPROGRESS , +and sends a SYN with a Fast Open cookie request automatically. +The caller needs to write the data again when the socket is connected. +On errors, +it sets the same +.I errno +as +.BR connect (2) +if the handshake fails. +This flag requires enabling TCP Fast Open client support on sysctl +.IR net.ipv4.tcp_fastopen . +.IP +Refer to +.B TCP_FASTOPEN_CONNECT +socket option in +.BR tcp (7) +for an alternative approach. +.SS sendmsg() +The definition of the +.I msghdr +structure employed by +.BR sendmsg () +is as follows: +.PP +.in +4n +.EX +struct msghdr { + void *msg_name; /* Optional address */ + socklen_t msg_namelen; /* Size of address */ + struct iovec *msg_iov; /* Scatter/gather array */ + size_t msg_iovlen; /* # elements in msg_iov */ + void *msg_control; /* Ancillary data, see below */ + size_t msg_controllen; /* Ancillary data buffer len */ + int msg_flags; /* Flags (unused) */ +}; +.EE +.in +.PP +The +.I msg_name +field is used on an unconnected socket to specify the target +address for a datagram. +It points to a buffer containing the address; the +.I msg_namelen +field should be set to the size of the address. +For a connected socket, these fields should be specified as NULL and 0, +respectively. +.PP +The +.I msg_iov +and +.I msg_iovlen +fields specify scatter-gather locations, as for +.BR writev (2). +.PP +You may send control information (ancillary data) using the +.I msg_control +and +.I msg_controllen +members. +The maximum control buffer length the kernel can process is limited +per socket by the value in +.IR /proc/sys/net/core/optmem_max ; +see +.BR socket (7). +For further information on the use of ancillary data in various +socket domains, see +.BR unix (7) +and +.BR ip (7). +.PP +The +.I msg_flags +field is ignored. +.\" Still to be documented: +.\" Send file descriptors and user credentials using the +.\" msg_control* fields. +.SH RETURN VALUE +On success, these calls return the number of bytes sent. +On error, \-1 is returned, and +.I errno +is set to indicate the error. +.SH ERRORS +These are some standard errors generated by the socket layer. +Additional errors +may be generated and returned from the underlying protocol modules; +see their respective manual pages. +.TP +.B EACCES +(For UNIX domain sockets, which are identified by pathname) +Write permission is denied on the destination socket file, +or search permission is denied for one of the directories +the path prefix. +(See +.BR path_resolution (7).) +.IP +(For UDP sockets) An attempt was made to send to a +network/broadcast address as though it was a unicast address. +.TP +.BR EAGAIN " or " EWOULDBLOCK +.\" Actually EAGAIN on Linux +The socket is marked nonblocking and the requested operation +would block. +POSIX.1-2001 allows either error to be returned for this case, +and does not require these constants to have the same value, +so a portable application should check for both possibilities. +.TP +.B EAGAIN +(Internet domain datagram sockets) +The socket referred to by +.I sockfd +had not previously been bound to an address and, +upon attempting to bind it to an ephemeral port, +it was determined that all port numbers in the ephemeral port range +are currently in use. +See the discussion of +.I /proc/sys/net/ipv4/ip_local_port_range +in +.BR ip (7). +.TP +.B EALREADY +Another Fast Open is in progress. +.TP +.B EBADF +.I sockfd +is not a valid open file descriptor. +.TP +.B ECONNRESET +Connection reset by peer. +.TP +.B EDESTADDRREQ +The socket is not connection-mode, and no peer address is set. +.TP +.B EFAULT +An invalid user space address was specified for an argument. +.TP +.B EINTR +A signal occurred before any data was transmitted; see +.BR signal (7). +.TP +.B EINVAL +Invalid argument passed. +.TP +.B EISCONN +The connection-mode socket was connected already but a +recipient was specified. +(Now either this error is returned, or the recipient specification +is ignored.) +.TP +.B EMSGSIZE +The socket type +.\" (e.g., SOCK_DGRAM ) +requires that message be sent atomically, and the size +of the message to be sent made this impossible. +.TP +.B ENOBUFS +The output queue for a network interface was full. +This generally indicates that the interface has stopped sending, +but may be caused by transient congestion. +(Normally, this does not occur in Linux. +Packets are just silently dropped +when a device queue overflows.) +.TP +.B ENOMEM +No memory available. +.TP +.B ENOTCONN +The socket is not connected, and no target has been given. +.TP +.B ENOTSOCK +The file descriptor +.I sockfd +does not refer to a socket. +.TP +.B EOPNOTSUPP +Some bit in the +.I flags +argument is inappropriate for the socket type. +.TP +.B EPIPE +The local end has been shut down on a connection oriented socket. +In this case, the process +will also receive a +.B SIGPIPE +unless +.B MSG_NOSIGNAL +is set. +.SH VERSIONS +According to POSIX.1-2001, the +.I msg_controllen +field of the +.I msghdr +structure should be typed as +.IR socklen_t , +and the +.I msg_iovlen +field should be typed as +.IR int , +but glibc currently types both as +.IR size_t . +.\" glibc bug for msg_controllen raised 12 Mar 2006 +.\" http://sourceware.org/bugzilla/show_bug.cgi?id=2448 +.\" The problem is an underlying kernel issue: the size of the +.\" __kernel_size_t type used to type these fields varies +.\" across architectures, but socklen_t is always 32 bits, +.\" as (at least with GCC) is int. +.SH STANDARDS +POSIX.1-2008. +.PP +.B MSG_CONFIRM +is a Linux extension. +.SH HISTORY +4.4BSD, SVr4, POSIX.1-2001. +(first appeared in 4.2BSD). +.PP +POSIX.1-2001 describes only the +.B MSG_OOB +and +.B MSG_EOR +flags. +POSIX.1-2008 adds a specification of +.BR MSG_NOSIGNAL . +.SH NOTES +See +.BR sendmmsg (2) +for information about a Linux-specific system call +that can be used to transmit multiple datagrams in a single call. +.SH BUGS +Linux may return +.B EPIPE +instead of +.BR ENOTCONN . +.SH EXAMPLES +An example of the use of +.BR sendto () +is shown in +.BR getaddrinfo (3). +.SH SEE ALSO +.BR fcntl (2), +.BR getsockopt (2), +.BR recv (2), +.BR select (2), +.BR sendfile (2), +.BR sendmmsg (2), +.BR shutdown (2), +.BR socket (2), +.BR write (2), +.BR cmsg (3), +.BR ip (7), +.BR ipv6 (7), +.BR socket (7), +.BR tcp (7), +.BR udp (7), +.BR unix (7) |