man7/socket.7

   1 '\" t
   2 .\" Don't change the first line, it tells man that we need tbl.
   3 .\" This man page is Copyright (C) 1999 Andi Kleen <ak@muc.de>.
   4 .\" and copyright (c) 1999 Matthew Wilcox.
   5 .\" Permission is granted to distribute possibly modified copies
   6 .\" of this page provided the header is included verbatim,
   7 .\" and in case of nontrivial modification author and date
   8 .\" of the modification is added to the header.
   9 .\"
  10 .\" 2002-10-30, Michael Kerrisk, <mtk.manpages@gmail.com>
  11 .\"     Added description of SO_ACCEPTCONN
  12 .\" 2004-05-20, aeb, added SO_RCVTIMEO/SO_SNDTIMEO text.
  13 .\" Modified, 27 May 2004, Michael Kerrisk <mtk.manpages@gmail.com>
  14 .\"     Added notes on capability requirements
  15 .\"     A few small grammar fixes
  16 .\" 2010-06-13 Jan Engelhardt <jengelh@medozas.de>
  17 .\"     Documented SO_DOMAIN and SO_PROTOCOL.
  18 .\" FIXME
  19 .\" The following are not yet documented:
  20 .\"     SO_PEERNAME (2.4?)
  21 .\"             get only
  22 .\"             Seems to do something similar to getpeernam(), but then
  23 .\"             why is it necessary / how does it differ?
  24 .\"     SO_TIMESTAMPNS (2.6.22)
  25 .\"             Documentation/networking/timestamping.txt
  26 .\"             commit 92f37fd2ee805aa77925c1e64fd56088b46094fc
  27 .\"             Author: Eric Dumazet <dada1@cosmosbay.com>
  28 .\"     SO_TIMESTAMPING (2.6.30)
  29 .\"             Documentation/networking/timestamping.txt
  30 .\"             commit cb9eff097831007afb30d64373f29d99825d0068
  31 .\"             Author: Patrick Ohly <patrick.ohly@intel.com>
  32 .\"     SO_RXQ_OVFL (2.6.33)
  33 .\"             commit 3b885787ea4112eaa80945999ea0901bf742707f
  34 .\"             Author: Neil Horman <nhorman@tuxdriver.com>
  35 .\"     SO_WIFI_STATUS (3.3)
  36 .\"             commit 6e3e939f3b1bf8534b32ad09ff199d88800835a0
  37 .\"             Author: Johannes Berg <johannes.berg@intel.com>
  38 .\"             Also: SCM_WIFI_STATUS
  39 .\"     SO_PEEK_OFF (3.4)
  40 .\"             commit ef64a54f6e558155b4f149bb10666b9e914b6c54
  41 .\"             Author: Pavel Emelyanov <xemul@parallels.com>
  42 .\"     SO_NOFCS (3.4)
  43 .\"             commit 3bdc0eba0b8b47797f4a76e377dd8360f317450f
  44 .\"             Author: Ben Greear <greearb@candelatech.com>
  45 .\"
  46 .TH SOCKET 7 2012-12-15 Linux "Linux Programmer's Manual"
  47 .SH NAME
  48 socket \- Linux socket interface
  49 .SH SYNOPSIS
  50 .B #include <sys/socket.h>
  51 .sp
  52 .IB sockfd " = socket(int " socket_family ", int " socket_type ", int " protocol );
  53 .SH DESCRIPTION
  54 This manual page describes the Linux networking socket layer user
  55 interface.
  56 The BSD compatible sockets
  57 are the uniform interface
  58 between the user process and the network protocol stacks in the kernel.
  59 The protocol modules are grouped into
  60 .I protocol families
  61 like
  62 .BR AF_INET ", " AF_IPX ", " AF_PACKET
  63 and
  64 .I socket types
  65 like
  66 .B SOCK_STREAM
  67 or
  68 .BR SOCK_DGRAM .
  69 See
  70 .BR socket (2)
  71 for more information on families and types.
  72 .SS Socket-layer functions
  73 These functions are used by the user process to send or receive packets
  74 and to do other socket operations.
  75 For more information see their respective manual pages.
  76
  77 .BR socket (2)
  78 creates a socket,
  79 .BR connect (2)
  80 connects a socket to a remote socket address,
  81 the
  82 .BR bind (2)
  83 function binds a socket to a local socket address,
  84 .BR listen (2)
  85 tells the socket that new connections shall be accepted, and
  86 .BR accept (2)
  87 is used to get a new socket with a new incoming connection.
  88 .BR socketpair (2)
  89 returns two connected anonymous sockets (only implemented for a few
  90 local families like
  91 .BR AF_UNIX )
  92 .PP
  93 .BR send (2),
  94 .BR sendto (2),
  95 and
  96 .BR sendmsg (2)
  97 send data over a socket, and
  98 .BR recv (2),
  99 .BR recvfrom (2),
 100 .BR recvmsg (2)
 101 receive data from a socket.
 102 .BR poll (2)
 103 and
 104 .BR select (2)
 105 wait for arriving data or a readiness to send data.
 106 In addition, the standard I/O operations like
 107 .BR write (2),
 108 .BR writev (2),
 109 .BR sendfile (2),
 110 .BR read (2),
 111 and
 112 .BR readv (2)
 113 can be used to read and write data.
 114 .PP
 115 .BR getsockname (2)
 116 returns the local socket address and
 117 .BR getpeername (2)
 118 returns the remote socket address.
 119 .BR getsockopt (2)
 120 and
 121 .BR setsockopt (2)
 122 are used to set or get socket layer or protocol options.
 123 .BR ioctl (2)
 124 can be used to set or read some other options.
 125 .PP
 126 .BR close (2)
 127 is used to close a socket.
 128 .BR shutdown (2)
 129 closes parts of a full-duplex socket connection.
 130 .PP
 131 Seeking, or calling
 132 .BR pread (2)
 133 or
 134 .BR pwrite (2)
 135 with a nonzero position is not supported on sockets.
 136 .PP
 137 It is possible to do nonblocking I/O on sockets by setting the
 138 .B O_NONBLOCK
 139 flag on a socket file descriptor using
 140 .BR fcntl (2).
 141 Then all operations that would block will (usually)
 142 return with
 143 .B EAGAIN
 144 (operation should be retried later);
 145 .BR connect (2)
 146 will return
 147 .B EINPROGRESS
 148 error.
 149 The user can then wait for various events via
 150 .BR poll (2)
 151 or
 152 .BR select (2).
 153 .TS
 154 tab(:) allbox;
 155 c s s
 156 l l l.
 157 I/O events
 158 Event:Poll flag:Occurrence
 159 Read:POLLIN:T{
 160 New data arrived.
 161 T}
 162 Read:POLLIN:T{
 163 A connection setup has been completed
 164 (for connection-oriented sockets)
 165 T}
 166 Read:POLLHUP:T{
 167 A disconnection request has been initiated by the other end.
 168 T}
 169 Read:POLLHUP:T{
 170 A connection is broken (only for connection-oriented protocols).
 171 When the socket is written
 172 .B SIGPIPE
 173 is also sent.
 174 T}
 175 Write:POLLOUT:T{
 176 Socket has enough send buffer space for writing new data.
 177 T}
 178 Read/Write:T{
 179 POLLIN|
 180 .br
 181 POLLOUT
 182 T}:T{
 183 An outgoing
 184 .BR connect (2)
 185 finished.
 186 T}
 187 Read/Write:POLLERR:An asynchronous error occurred.
 188 Read/Write:POLLHUP:The other end has shut down one direction.
 189 Exception:POLLPRI:T{
 190 Urgent data arrived.
 191 .B SIGURG
 192 is sent then.
 193 T}
 194 .\" FIXME . The following is not true currently:
 195 .\" It is no I/O event when the connection
 196 .\" is broken from the local end using
 197 .\" .BR shutdown (2)
 198 .\" or
 199 .\" .BR close (2).
 200 .TE
 201 .PP
 202 An alternative to
 203 .BR poll (2)
 204 and
 205 .BR select (2)
 206 is to let the kernel inform the application about events
 207 via a
 208 .B SIGIO
 209 signal.
 210 For that the
 211 .B O_ASYNC
 212 flag must be set on a socket file descriptor via
 213 .BR fcntl (2)
 214 and a valid signal handler for
 215 .B SIGIO
 216 must be installed via
 217 .BR sigaction (2).
 218 See the
 219 .I Signals
 220 discussion below.
 221 .SS Socket address structures
 222 Each socket domain has its own format for socket addresses,
 223 with a domain-specific address structure.
 224 Each of these structures begins with an
 225 integer "family" field (typed as
 226 .IR sa_family_t )
 227 that indicates the type of the address structure.
 228 This allows
 229 the various system calls (e.g.,
 230 .BR connect (2),
 231 .BR bind (2),
 232 .BR accept (2),
 233 .BR getsockname (2),
 234 .BR getpeername (2)),
 235 which are generic to all socket domains,
 236 to determine the domain of a particular socket address.
 237
 238 To allow any type of socket address to be passed to
 239 interfaces in the sockets API,
 240 the type
 241 .IR "struct sockaddr"
 242 is defined.
 243 The purpose of this type is purely to allow casting of
 244 domain-specific socket address types to a "generic" type,
 245 so as to avoid compiler warnings about type mismatches in
 246 calls to the sockets API.
 247
 248 In addition, the sockets API provides the data type
 249 .IR "struct sockaddr_storage".
 250 This type
 251 is suitable to accommodate all supported domain-specific socket
 252 address structures; it is large enough and is aligned properly.
 253 (In particular, it is large enough to hold
 254 IPv6 socket addresses.)
 255 The structure includes the following field, which can be used to identify
 256 the type of socket address actually stored in the structure:
 257
 258 .in +4n
 259 .nf
 260     sa_family_t ss_family;
 261 .fi
 262 .in
 263
 264 The
 265 .I sockaddr_storage
 266 structure is useful in programs that must handle socket addresses
 267 in a generic way
 268 (e.g., programs that must deal with both IPv4 and IPv6 socket addresses).
 269 .SS Socket options
 270 These socket options can be set by using
 271 .BR setsockopt (2)
 272 and read with
 273 .BR getsockopt (2)
 274 with the socket level set to
 275 .B SOL_SOCKET
 276 for all sockets:
 277 .\" FIXME
 278 .\" In the list below, the text used to describe argument types
 279 .\" for each socket option should be more consistent
 280 .\"
 281 .\" SO_ACCEPTCONN is in POSIX.1-2001, and its origin is explained in
 282 .\" W R Stevens, UNPv1
 283 .TP
 284 .B SO_ACCEPTCONN
 285 Returns a value indicating whether or not this socket has been marked
 286 to accept connections with
 287 .BR listen (2).
 288 The value 0 indicates that this is not a listening socket,
 289 the value 1 indicates that this is a listening socket.
 290 This socket option is read-only.
 291 .TP
 292 .B SO_BINDTODEVICE
 293 Bind this socket to a particular device like \(lqeth0\(rq,
 294 as specified in the passed interface name.
 295 If the
 296 name is an empty string or the option length is zero, the socket device
 297 binding is removed.
 298 The passed option is a variable-length null-terminated
 299 interface name string with the maximum size of
 300 .BR IFNAMSIZ .
 301 If a socket is bound to an interface,
 302 only packets received from that particular interface are processed by the
 303 socket.
 304 Note that this only works for some socket types, particularly
 305 .B AF_INET
 306 sockets.
 307 It is not supported for packet sockets (use normal
 308 .BR bind (2)
 309 there).
 310
 311 This option is available for getting since Linux 3.8. The option
 312 length should contain the buffer size to receive the device name
 313 and is recommended to be
 314 .BR IFNAMSZ
 315 bytes. The real device name length is reported back in the option
 316 length argument.
 317 .TP
 318 .B SO_BROADCAST
 319 Set or get the broadcast flag.
 320 When enabled, datagram sockets are allowed to send
 321 packets to a broadcast address.
 322 This option has no effect on stream-oriented sockets.
 323 .TP
 324 .B SO_BSDCOMPAT
 325 Enable BSD bug-to-bug compatibility.
 326 This is used by the UDP protocol module in Linux 2.0 and 2.2.
 327 If enabled ICMP errors received for a UDP socket will not be passed
 328 to the user program.
 329 In later kernel versions, support for this option has been phased out:
 330 Linux 2.4 silently ignores it, and Linux 2.6 generates a kernel warning
 331 (printk()) if a program uses this option.
 332 Linux 2.0 also enabled BSD bug-to-bug compatibility
 333 options (random header changing, skipping of the broadcast flag) for raw
 334 sockets with this option, but that was removed in Linux 2.2.
 335 .TP
 336 .B SO_DEBUG
 337 Enable socket debugging.
 338 Only allowed for processes with the
 339 .B CAP_NET_ADMIN
 340 capability or an effective user ID of 0.
 341 .TP
 342 .BR SO_DOMAIN " (since Linux 2.6.32)"
 343 Retrieves the socket domain as an integer, returning a value such as
 344 .BR AF_INET6 .
 345 See
 346 .BR socket (2)
 347 for details.
 348 This socket option is read-only.
 349 .TP
 350 .B SO_ERROR
 351 Get and clear the pending socket error.
 352 This socket option is read-only.
 353 Expects an integer.
 354 .TP
 355 .B SO_DONTROUTE
 356 Don't send via a gateway, only send to directly connected hosts.
 357 The same effect can be achieved by setting the
 358 .B MSG_DONTROUTE
 359 flag on a socket
 360 .BR send (2)
 361 operation.
 362 Expects an integer boolean flag.
 363 .TP
 364 .B SO_KEEPALIVE
 365 Enable sending of keep-alive messages on connection-oriented sockets.
 366 Expects an integer boolean flag.
 367 .TP
 368 .B SO_LINGER
 369 Sets or gets the
 370 .B SO_LINGER
 371 option.
 372 The argument is a
 373 .I linger
 374 structure.
 375 .sp
 376 .in +4n
 377 .nf
 378 struct linger {
 379     int l_onoff;    /* linger active */
 380     int l_linger;   /* how many seconds to linger for */
 381 };
 382 .fi
 383 .in
 384 .IP
 385 When enabled, a
 386 .BR close (2)
 387 or
 388 .BR shutdown (2)
 389 will not return until all queued messages for the socket have been
 390 successfully sent or the linger timeout has been reached.
 391 Otherwise,
 392 the call returns immediately and the closing is done in the background.
 393 When the socket is closed as part of
 394 .BR exit (2),
 395 it always lingers in the background.
 396 .TP
 397 .BR SO_MARK " (since Linux 2.6.25)"
 398 .\" commit 4a19ec5800fc3bb64e2d87c4d9fdd9e636086fe0
 399 .\" and    914a9ab386a288d0f22252fc268ecbc048cdcbd5
 400 Set the mark for each packet sent through this socket
 401 (similar to the netfilter MARK target but socket-based).
 402 Changing the mark can be used for mark-based
 403 routing without netfilter or for packet filtering.
 404 Setting this option requires the
 405 .B CAP_NET_ADMIN
 406 capability.
 407 .TP
 408 .B SO_OOBINLINE
 409 If this option is enabled,
 410 out-of-band data is directly placed into the receive data stream.
 411 Otherwise out-of-band data is only passed when the
 412 .B MSG_OOB
 413 flag is set during receiving.
 414 .\" don't document it because it can do too much harm.
 415 .\".B SO_NO_CHECK
 416 .TP
 417 .B SO_PASSCRED
 418 Enable or disable the receiving of the
 419 .B SCM_CREDENTIALS
 420 control message.
 421 For more information see
 422 .BR unix (7).
 423 .\" FIXME Document SO_PASSSEC, added in 2.6.18; there is some info
 424 .\" in the 2.6.18 ChangeLog
 425 .TP
 426 .B SO_PEERCRED
 427 Return the credentials of the foreign process connected to this socket.
 428 This is only possible for connected
 429 .B AF_UNIX
 430 stream sockets and
 431 .B AF_UNIX
 432 stream and datagram socket pairs created using
 433 .BR socketpair (2);
 434 see
 435 .BR unix (7).
 436 The returned credentials are those that were in effect at the time
 437 of the call to
 438 .BR connect (2)
 439 or
 440 .BR socketpair (2).
 441 Argument is a
 442 .I ucred
 443 structure.
 444 This socket option is read-only.
 445 .TP
 446 .B SO_PRIORITY
 447 Set the protocol-defined priority for all packets to be sent on
 448 this socket.
 449 Linux uses this value to order the networking queues:
 450 packets with a higher priority may be processed first depending
 451 on the selected device queueing discipline.
 452 For
 453 .BR ip (7),
 454 this also sets the IP type-of-service (TOS) field for outgoing packets.
 455 Setting a priority outside the range 0 to 6 requires the
 456 .B CAP_NET_ADMIN
 457 capability.
 458 .TP
 459 .BR SO_PROTOCOL " (since Linux 2.6.32)"
 460 Retrieves the socket protocol as an integer, returning a value such as
 461 .BR IPPROTO_SCTP .
 462 See
 463 .BR socket (2)
 464 for details.
 465 This socket option is read-only.
 466 .TP
 467 .B SO_RCVBUF
 468 Sets or gets the maximum socket receive buffer in bytes.
 469 The kernel doubles this value (to allow space for bookkeeping overhead)
 470 when it is set using
 471 .\" Most (all?) other implementations do not do this -- MTK, Dec 05
 472 .BR setsockopt (2),
 473 and this doubled value is returned by
 474 .BR getsockopt (2).
 475 .\" The following thread on LMKL is quite informative:
 476 .\" getsockopt/setsockopt with SO_RCVBUF and SO_SNDBUF "non-standard" behaviour
 477 .\" 17 July 2012
 478 .\" http://thread.gmane.org/gmane.linux.kernel/1328935
 479 The default value is set by the
 480 .I /proc/sys/net/core/rmem_default
 481 file, and the maximum allowed value is set by the
 482 .I /proc/sys/net/core/rmem_max
 483 file.
 484 The minimum (doubled) value for this option is 256.
 485 .TP
 486 .BR SO_RCVBUFFORCE " (since Linux 2.6.14)"
 487 Using this socket option, a privileged
 488 .RB ( CAP_NET_ADMIN )
 489 process can perform the same task as
 490 .BR SO_RCVBUF ,
 491 but the
 492 .I rmem_max
 493 limit can be overridden.
 494 .TP
 495 .BR SO_RCVLOWAT " and " SO_SNDLOWAT
 496 Specify the minimum number of bytes in the buffer until the socket layer
 497 will pass the data to the protocol
 498 .RB ( SO_SNDLOWAT )
 499 or the user on receiving
 500 .RB ( SO_RCVLOWAT ).
 501 These two values are initialized to 1.
 502 .B SO_SNDLOWAT
 503 is not changeable on Linux
 504 .RB ( setsockopt (2)
 505 fails with the error
 506 .BR ENOPROTOOPT ).
 507 .B SO_RCVLOWAT
 508 is changeable
 509 only since Linux 2.4.
 510 The
 511 .BR select (2)
 512 and
 513 .BR poll (2)
 514 system calls currently do not respect the
 515 .B SO_RCVLOWAT
 516 setting on Linux,
 517 and mark a socket readable when even a single byte of data is available.
 518 A subsequent read from the socket will block until
 519 .B SO_RCVLOWAT
 520 bytes are available.
 521 .\" See http://marc.theaimsgroup.com/?l=linux-kernel&m=111049368106984&w=2
 522 .\" Tested on kernel 2.6.14 -- mtk, 30 Nov 05
 523 .TP
 524 .BR SO_RCVTIMEO " and " SO_SNDTIMEO
 525 .\" Not implemented in 2.0.
 526 .\" Implemented in 2.1.11 for getsockopt: always return a zero struct.
 527 .\" Implemented in 2.3.41 for setsockopt, and actually used.
 528 Specify the receiving or sending timeouts until reporting an error.
 529 The argument is a
 530 .IR "struct timeval" .
 531 If an input or output function blocks for this period of time, and
 532 data has been sent or received, the return value of that function
 533 will be the amount of data transferred; if no data has been transferred
 534 and the timeout has been reached then \-1 is returned with
 535 .I errno
 536 set to
 537 .BR EAGAIN
 538 or
 539 .BR EWOULDBLOCK ,
 540 .\" in fact to EAGAIN
 541 or
 542 .B EINPROGRESS
 543 (for
 544 .BR connect (2))
 545 just as if the socket was specified to be nonblocking.
 546 If the timeout is set to zero (the default)
 547 then the operation will never timeout.
 548 Timeouts only have effect for system calls that perform socket I/O (e.g.,
 549 .BR read (2),
 550 .BR recvmsg (2),
 551 .BR send (2),
 552 .BR sendmsg (2));
 553 timeouts have no effect for
 554 .BR select (2),
 555 .BR poll (2),
 556 .BR epoll_wait (2),
 557 and so on.
 558 .TP
 559 .B SO_REUSEADDR
 560 Indicates that the rules used in validating addresses supplied in a
 561 .BR bind (2)
 562 call should allow reuse of local addresses.
 563 For
 564 .B AF_INET
 565 sockets this
 566 means that a socket may bind, except when there
 567 is an active listening socket bound to the address.
 568 When the listening socket is bound to
 569 .B INADDR_ANY
 570 with a specific port then it is not possible
 571 to bind to this port for any local address.
 572 Argument is an integer boolean flag.
 573 .TP
 574 .B SO_SNDBUF
 575 Sets or gets the maximum socket send buffer in bytes.
 576 The kernel doubles this value (to allow space for bookkeeping overhead)
 577 when it is set using
 578 .\" Most (all?) other implementations do not do this -- MTK, Dec 05
 579 .\" See also the comment to SO_RCVBUF (17 Jul 2012 LKML mail)
 580 .BR setsockopt (2),
 581 and this doubled value is returned by
 582 .BR getsockopt (2).
 583 The default value is set by the
 584 .I /proc/sys/net/core/wmem_default
 585 file and the maximum allowed value is set by the
 586 .I /proc/sys/net/core/wmem_max
 587 file.
 588 The minimum (doubled) value for this option is 2048.
 589 .TP
 590 .BR SO_SNDBUFFORCE " (since Linux 2.6.14)"
 591 Using this socket option, a privileged
 592 .RB ( CAP_NET_ADMIN )
 593 process can perform the same task as
 594 .BR SO_SNDBUF ,
 595 but the
 596 .I wmem_max
 597 limit can be overridden.
 598 .TP
 599 .B SO_TIMESTAMP
 600 Enable or disable the receiving of the
 601 .B SO_TIMESTAMP
 602 control message.
 603 The timestamp control message is sent with level
 604 .B SOL_SOCKET
 605 and the
 606 .I cmsg_data
 607 field is a
 608 .I "struct timeval"
 609 indicating the
 610 reception time of the last packet passed to the user in this call.
 611 See
 612 .BR cmsg (3)
 613 for details on control messages.
 614 .TP
 615 .B SO_TYPE
 616 Gets the socket type as an integer (e.g.,
 617 .BR SOCK_STREAM ).
 618 This socket option is read-only.
 619 .SS Signals
 620 When writing onto a connection-oriented socket that has been shut down
 621 (by the local or the remote end)
 622 .B SIGPIPE
 623 is sent to the writing process and
 624 .B EPIPE
 625 is returned.
 626 The signal is not sent when the write call
 627 specified the
 628 .B MSG_NOSIGNAL
 629 flag.
 630 .PP
 631 When requested with the
 632 .B FIOSETOWN
 633 .BR fcntl (2)
 634 or
 635 .B SIOCSPGRP
 636 .BR ioctl (2),
 637 .B SIGIO
 638 is sent when an I/O event occurs.
 639 It is possible to use
 640 .BR poll (2)
 641 or
 642 .BR select (2)
 643 in the signal handler to find out which socket the event occurred on.
 644 An alternative (in Linux 2.2) is to set a real-time signal using the
 645 .B F_SETSIG
 646 .BR fcntl (2);
 647 the handler of the real time signal will be called with
 648 the file descriptor in the
 649 .I si_fd
 650 field of its
 651 .IR siginfo_t .
 652 See
 653 .BR fcntl (2)
 654 for more information.
 655 .PP
 656 Under some circumstances (e.g., multiple processes accessing a
 657 single socket), the condition that caused the
 658 .B SIGIO
 659 may have already disappeared when the process reacts to the signal.
 660 If this happens, the process should wait again because Linux
 661 will resend the signal later.
 662 .\" .SS Ancillary messages
 663 .SS /proc interfaces
 664 The core socket networking parameters can be accessed
 665 via files in the directory
 666 .IR /proc/sys/net/core/ .
 667 .TP
 668 .I rmem_default
 669 contains the default setting in bytes of the socket receive buffer.
 670 .TP
 671 .I rmem_max
 672 contains the maximum socket receive buffer size in bytes which a user may
 673 set by using the
 674 .B SO_RCVBUF
 675 socket option.
 676 .TP
 677 .I wmem_default
 678 contains the default setting in bytes of the socket send buffer.
 679 .TP
 680 .I wmem_max
 681 contains the maximum socket send buffer size in bytes which a user may
 682 set by using the
 683 .B SO_SNDBUF
 684 socket option.
 685 .TP
 686 .IR message_cost " and " message_burst
 687 configure the token bucket filter used to load limit warning messages
 688 caused by external network events.
 689 .TP
 690 .I netdev_max_backlog
 691 Maximum number of packets in the global input queue.
 692 .TP
 693 .I optmem_max
 694 Maximum length of ancillary data and user control data like the iovecs
 695 per socket.
 696 .\" netdev_fastroute is not documented because it is experimental
 697 .SS Ioctls
 698 These operations can be accessed using
 699 .BR ioctl (2):
 700
 701 .in +4n
 702 .nf
 703 .IB error " = ioctl(" ip_socket ", " ioctl_type ", " &value_result ");"
 704 .fi
 705 .in
 706 .TP
 707 .B SIOCGSTAMP
 708 Return a
 709 .I struct timeval
 710 with the receive timestamp of the last packet passed to the user.
 711 This is useful for accurate round trip time measurements.
 712 See
 713 .BR setitimer (2)
 714 for a description of
 715 .IR "struct timeval" .
 716 .\"
 717 This ioctl should only be used if the socket option
 718 .B SO_TIMESTAMP
 719 is not set on the socket.
 720 Otherwise, it returns the timestamp of the
 721 last packet that was received while
 722 .B SO_TIMESTAMP
 723 was not set, or it fails if no such packet has been received,
 724 (i.e.,
 725 .BR ioctl (2)
 726 returns \-1 with
 727 .I errno
 728 set to
 729 .BR ENOENT ).
 730 .TP
 731 .B SIOCSPGRP
 732 Set the process or process group to send
 733 .B SIGIO
 734 or
 735 .B SIGURG
 736 signals
 737 to when an
 738 asynchronous I/O operation has finished or urgent data is available.
 739 The argument is a pointer to a
 740 .IR pid_t .
 741 If the argument is positive, send the signals to that process.
 742 If the
 743 argument is negative, send the signals to the process group with the ID
 744 of the absolute value of the argument.
 745 The process may only choose itself or its own process group to receive
 746 signals unless it has the
 747 .B CAP_KILL
 748 capability or an effective UID of 0.
 749 .TP
 750 .B FIOASYNC
 751 Change the
 752 .B O_ASYNC
 753 flag to enable or disable asynchronous I/O mode of the socket.
 754 Asynchronous I/O mode means that the
 755 .B SIGIO
 756 signal or the signal set with
 757 .B F_SETSIG
 758 is raised when a new I/O event occurs.
 759 .IP
 760 Argument is an integer boolean flag.
 761 (This operation is synonymous with the use of
 762 .BR fcntl (2)
 763 to set the
 764 .B O_ASYNC
 765 flag.)
 766 .\"
 767 .TP
 768 .B SIOCGPGRP
 769 Get the current process or process group that receives
 770 .B SIGIO
 771 or
 772 .B SIGURG
 773 signals,
 774 or 0
 775 when none is set.
 776 .PP
 777 Valid
 778 .BR fcntl (2)
 779 operations:
 780 .TP
 781 .B FIOGETOWN
 782 The same as the
 783 .B SIOCGPGRP
 784 .BR ioctl (2).
 785 .TP
 786 .B FIOSETOWN
 787 The same as the
 788 .B SIOCSPGRP
 789 .BR ioctl (2).
 790 .SH VERSIONS
 791 .B SO_BINDTODEVICE
 792 was introduced in Linux 2.0.30.
 793 .B SO_PASSCRED
 794 is new in Linux 2.2.
 795 The
 796 .I /proc
 797 interfaces was introduced in Linux 2.2.
 798 .B SO_RCVTIMEO
 799 and
 800 .B SO_SNDTIMEO
 801 are supported since Linux 2.3.41.
 802 Earlier, timeouts were fixed to
 803 a protocol-specific setting, and could not be read or written.
 804 .SH NOTES
 805 Linux assumes that half of the send/receive buffer is used for internal
 806 kernel structures; thus the values in the corresponding
 807 .I /proc
 808 files are twice what can be observed on the wire.
 809
 810 Linux will only allow port reuse with the
 811 .B SO_REUSEADDR
 812 option
 813 when this option was set both in the previous program that performed a
 814 .BR bind (2)
 815 to the port and in the program that wants to reuse the port.
 816 This differs from some implementations (e.g., FreeBSD)
 817 where only the later program needs to set the
 818 .B SO_REUSEADDR
 819 option.
 820 Typically this difference is invisible, since, for example, a server
 821 program is designed to always set this option.
 822 .SH BUGS
 823 The
 824 .B CONFIG_FILTER
 825 socket options
 826 .B SO_ATTACH_FILTER
 827 and
 828 .B SO_DETACH_FILTER
 829 .\" FIXME Document SO_ATTACH_FILTER and SO_DETACH_FILTER
 830 are not documented.
 831 The suggested interface to use them is via the libpcap
 832 library.
 833 .\" .SH AUTHORS
 834 .\" This man page was written by Andi Kleen.
 835 .SH SEE ALSO
 836 .BR getsockopt (2),
 837 .BR connect (2),
 838 .BR setsockopt (2),
 839 .BR socket (2),
 840 .BR capabilities (7),
 841 .BR ddp (7),
 842 .BR ip (7),
 843 .BR packet (7),
 844 .BR tcp (7),
 845 .BR udp (7),
 846 .BR unix (7)