linux accept4,linux - accept

ACCEPT(2)                  Linux Programmer's Manual                 ACCEPT(2)

NAME

accept - accept a connection on a socket

SYNOPSIS

#include /* See NOTES */

#include

int accept(int sockfd, struct sockaddr *addr, socklen_t *addrlen);

#define _GNU_SOURCE             /* See feature_test_macros(7) */

#include

int accept4(int sockfd, struct sockaddr *addr,

socklen_t *addrlen, int flags);

DESCRIPTION

The  accept()  system  call  is used with connection-based socket types

(SOCK_STREAM,  SOCK_SEQPACKET).   It  extracts  the  first   connection

request  on  the queue of pending connections for the listening socket,

sockfd, creates a new connected socket, and returns a new file descrip‐

tor  referring  to that socket.  The newly created socket is not in the

listening state.  The original socket  sockfd  is  unaffected  by  this

call.

The  argument  sockfd is a socket that has been created with socket(2),

bound to a local address with bind(2), and is listening for connections

after a listen(2).

The argument addr is a pointer to a sockaddr structure.  This structure

is filled in with the address of the peer socket, as known to the  com‐

munications  layer.   The  exact format of the address returned addr is

determined by the  socket's  address  family  (see  socket(2)  and  the

respective  protocol  man pages).  When addr is NULL, nothing is filled

in; in this case, addrlen is not used, and should also be NULL.

The addrlen argument is a value-result argument: the caller  must  ini‐

tialize  it  to contain the size (in bytes) of the structure pointed to

by addr; on return it will contain the actual size of the peer address.

The returned address is truncated if the buffer provided is too  small;

in  this case, addrlen will return a value greater than was supplied to

the call.

If no pending connections are present on the queue, and the  socket  is

not  marked  as nonblocking, accept() blocks the caller until a connec‐

tion is present.  If the socket is marked nonblocking  and  no  pending

connections  are  present  on  the queue, accept() fails with the error

EAGAIN or EWOULDBLOCK.

In order to be notified of incoming connections on a  socket,  you  can

use  select(2)  or  poll(2).  A readable event will be delivered when a

new connection is attempted and you may then call  accept()  to  get  a

socket  for  that connection.  Alternatively, you can set the socket to

deliver SIGIO when activity occurs  on  a  socket;  see  socket(7)  for

details.

For  certain  protocols which require an explicit confirmation, such as

DECNet, accept() can be thought of as merely dequeuing the next connec‐

tion  request  and  not  implying  confirmation.   Confirmation  can be

implied by a normal read or write  on  the  new  file  descriptor,  and

rejection  can  be  implied  by closing the new socket.  Currently only

DECNet has these semantics on Linux.

If flags is 0, then accept4() is the same as accept().   The  following

values can be bitwise ORed in flags to obtain different behavior:

SOCK_NONBLOCK   Set  the  O_NONBLOCK  file  status flag on the new open

file description.  Using this flag saves extra calls to

fcntl(2) to achieve the same result.

SOCK_CLOEXEC    Set the close-on-exec (FD_CLOEXEC) flag on the new file

descriptor.  See the description of the O_CLOEXEC  flag

in open(2) for reasons why this may be useful.

RETURN VALUE

On  success,  these system calls return a nonnegative integer that is a

descriptor for the accepted socket.  On  error,  -1  is  returned,  and

errno is set appropriately.

Error Handling

Linux accept() (and accept4()) passes already-pending network errors on

the new socket as an error code from accept().  This  behavior  differs

from  other  BSD  socket  implementations.   For reliable operation the

application should detect the network errors defined for  the  protocol

after  accept()  and  treat  them  like EAGAIN by retrying.  In case of

TCP/IP these are  ENETDOWN,  EPROTO,  ENOPROTOOPT,  EHOSTDOWN,  ENONET,

EHOSTUNREACH, EOPNOTSUPP, and ENETUNREACH.

ERRORS

EAGAIN or EWOULDBLOCK

The  socket is marked nonblocking and no connections are present

to be accepted.  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 pos‐

sibilities.

EBADF  The descriptor is invalid.

ECONNABORTED

A connection has been aborted.

EFAULT The  addr argument is not in a writable part of the user address

space.

EINTR  The system call was interrupted by  a  signal  that  was  caught

before a valid connection arrived; see signal(7).

EINVAL Socket  is  not listening for connections, or addrlen is invalid

(e.g., is negative).

EINVAL (accept4()) invalid value in flags.

EMFILE The per-process limit of open file descriptors has been reached.

ENFILE The system limit on the total number  of  open  files  has  been

reached.

ENOBUFS, ENOMEM

Not  enough free memory.  This often means that the memory allo‐

cation is limited by the socket buffer limits, not by the system

memory.

ENOTSOCK

The descriptor references a file, not a socket.

EOPNOTSUPP

The referenced socket is not of type SOCK_STREAM.

EPROTO Protocol error.

In addition, Linux accept() may fail if:

EPERM  Firewall rules forbid connection.

In  addition,  network errors for the new socket and as defined for the

protocol may be returned.   Various  Linux  kernels  can  return  other

errors such as ENOSR, ESOCKTNOSUPPORT, EPROTONOSUPPORT, ETIMEDOUT.  The

value ERESTARTSYS may be seen during a trace.

VERSIONS

The accept4() system call is available starting with Linux 2.6.28; sup‐

port in glibc is available starting with version 2.10.

CONFORMING TO

accept():  POSIX.1-2001,  SVr4,  4.4BSD,  (accept()  first  appeared in

4.2BSD).

accept4() is a nonstandard Linux extension.

On Linux, the new socket returned by accept()  does  not  inherit  file

status  flags such as O_NONBLOCK and O_ASYNC from the listening socket.

This behavior differs from the canonical  BSD  sockets  implementation.

Portable  programs  should not rely on inheritance or noninheritance of

file status flags and always explicitly set all required flags  on  the

socket returned from accept().

NOTES

POSIX.1-2001  does not require the inclusion of , and this

header file is not required on Linux.  However, some  historical  (BSD)

implementations  required  this  header file, and portable applications

are probably wise to include it.

There may not always be a connection waiting after a SIGIO is delivered

or  select(2) or poll(2) return a readability event because the connec‐

tion might have been  removed  by  an  asynchronous  network  error  or

another  thread  before  accept()  is called.  If this happens then the

call will block waiting for the next connection to arrive.   To  ensure

that  accept() never blocks, the passed socket sockfd needs to have the

O_NONBLOCK flag set (see socket(7)).

The socklen_t type

The third argument of accept() was originally declared as an int * (and

is  that  under libc4 and libc5 and on many other systems like 4.x BSD,

SunOS 4, SGI); a POSIX.1g draft standard wanted to  change  it  into  a

size_t  *, and that is what it is for SunOS 5.  Later POSIX drafts have

socklen_t *, and so do the Single Unix Specification and glibc2.  Quot‐

ing Linus Torvalds:

"_Any_  sane  library  _must_ have "socklen_t" be the same size as int.

Anything else breaks any BSD socket layer stuff.  POSIX  initially  did

make  it  a  size_t, and I (and hopefully others, but obviously not too

many) complained to them very loudly indeed.  Making  it  a  size_t  is

completely  broken, exactly because size_t very seldom is the same size

as "int" on 64-bit architectures, for example.  And it has  to  be  the

same  size  as  "int"  because that's what the BSD socket interface is.

Anyway,  the  POSIX  people  eventually  got  a   clue,   and   created

"socklen_t".   They  shouldn't  have touched it in the first place, but

once they did they felt it had to have a named type  for  some  unfath‐

omable  reason  (probably  somebody didn't like losing face over having

done the original stupid thing, so they  silently  just  renamed  their

blunder)."

EXAMPLE

See bind(2).

SEE ALSO

bind(2), connect(2), listen(2), select(2), socket(2), socket(7)

COLOPHON

This  page  is  part of release 3.27 of the Linux man-pages project.  A

description of the project, and information about reporting  bugs,  can

be found at .

Linux                             2010-09-10                         ACCEPT(2)