1 .\" Copyright (c) 1983, 1991, 1993
2 .\" The Regents of the University of California. All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
12 .\" 3. All advertising materials mentioning features or use of this software
13 .\" must display the following acknowledgement:
14 .\" This product includes software developed by the University of
15 .\" California, Berkeley and its contributors.
16 .\" 4. Neither the name of the University nor the names of its contributors
17 .\" may be used to endorse or promote products derived from this software
18 .\" without specific prior written permission.
20 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 .\" @(#)ip.4 8.2 (Berkeley) 11/30/93
46 .Fn socket AF_INET SOCK_RAW proto
49 is the transport layer protocol used
50 by the Internet protocol family.
51 Options may be set at the
54 when using higher-level protocols that are based on
60 It may also be accessed
63 when developing new protocols, or
64 special-purpose applications.
73 may be used to provide
75 options to be transmitted in the
77 header of each outgoing packet
78 or to examine the header options on incoming packets.
80 options may be used with any socket type in the Internet family.
83 options to be sent is that specified by the
85 protocol specification (RFC-791), with one exception:
86 the list of addresses for Source Route options must include the first-hop
87 gateway at the beginning of the list of gateways.
88 The first-hop gateway address will be extracted from the option list
89 and the size adjusted accordingly before use.
90 To disable previously specified options,
91 use a zero-length buffer:
93 setsockopt(s, IPPROTO_IP, IP_OPTIONS, NULL, 0);
99 may be used to set the type-of-service and time-to-live
103 .Dv SOCK_STREAM , SOCK_DGRAM ,
109 int tos = IPTOS_LOWDELAY; /* see <netinet/ip.h> */
110 setsockopt(s, IPPROTO_IP, IP_TOS, &tos, sizeof(tos));
112 int ttl = 60; /* max = 255 */
113 setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(ttl));
117 may be used to set the minimum acceptable TTL a packet must have when
118 received on a socket.
119 All packets with a lower TTL are silently dropped.
120 This option is only really useful when set to 255, preventing packets
121 from outside the directly connected networks reaching local listeners
125 may be used to set the Don't Fragment flag on IP packets.
126 Currently this option is respected only on
135 sockets, the Don't Fragment flag is controlled by the Path
136 MTU Discovery option.
137 Sending a packet larger than the MTU size of the egress interface,
138 determined by the destination address, returns an
144 option is enabled on a
149 call will return the destination
158 structure points to a buffer
161 structure followed by the
166 fields have the following values:
168 cmsg_len = CMSG_LEN(sizeof(struct in_addr))
169 cmsg_level = IPPROTO_IP
170 cmsg_type = IP_RECVDSTADDR
173 The source address to be used for outgoing
175 datagrams on a socket can be specified as ancillary data with a type code of
177 The msg_control field in the msghdr structure should point to a buffer
180 structure followed by the
183 The cmsghdr fields should have the following values:
185 cmsg_len = CMSG_LEN(sizeof(struct in_addr))
186 cmsg_level = IPPROTO_IP
187 cmsg_type = IP_SENDSRCADDR
190 The socket should be either bound to
192 and a local port, and the address supplied with
196 or the socket should be bound to a local address and the address supplied with
200 In the latter case bound address is overriden via generic source address
201 selection logic, which would choose IP address of interface closest to
206 is defined to have the same value as
212 can be used directly as a control message for
218 option is enabled on a
222 socket, the destination address of outgoing
223 broadcast datagrams on that socket will be forced
224 to the undirected broadcast address,
225 .Dv INADDR_BROADCAST ,
227 This is in contrast to the default behavior of the
228 system, which is to transmit undirected broadcasts
229 via the first network interface with the
233 This option allows applications to choose which
234 interface is used to transmit an undirected broadcast
236 For example, the following code would force an
237 undirected broadcast to be transmitted via the interface
238 configured with the broadcast address 192.168.2.255:
241 struct sockaddr_in sin;
242 int onesbcast = 1; /* 0 = disable (default), 1 = enable */
244 setsockopt(s, IPPROTO_IP, IP_ONESBCAST, &onesbcast, sizeof(onesbcast));
245 sin.sin_addr.s_addr = inet_addr("192.168.2.255");
246 sin.sin_port = htons(1234);
247 sendto(s, msg, sizeof(msg), 0, &sin, sizeof(sin));
250 It is the application's responsibility to set the
253 to an appropriate value in order to prevent broadcast storms.
254 The application must have sufficient credentials to set the
256 socket level option, otherwise the
258 option has no effect.
262 option is enabled on a
269 to any address, even one not bound to any available network interface in the
271 This functionality (in conjunction with special firewall rules) can be used for
272 implementing a transparent proxy.
274 .Dv PRIV_NETINET_BINDANY
275 privilege is needed to set this option.
279 option is enabled on a
286 (time to live) field for a
289 The msg_control field in the msghdr structure points to a buffer
290 that contains a cmsghdr structure followed by the
292 The cmsghdr fields have the following values:
294 cmsg_len = CMSG_LEN(sizeof(u_char))
295 cmsg_level = IPPROTO_IP
296 cmsg_type = IP_RECVTTL
302 option is enabled on a
309 (type of service) field for a
312 The msg_control field in the msghdr structure points to a buffer
313 that contains a cmsghdr structure followed by the
315 The cmsghdr fields have the following values:
317 cmsg_len = CMSG_LEN(sizeof(u_char))
318 cmsg_level = IPPROTO_IP
319 cmsg_type = IP_RECVTOS
325 option is enabled on a
330 .Vt "struct sockaddr_dl"
331 corresponding to the interface on which the
337 structure points to a buffer that contains a
339 structure followed by the
340 .Vt "struct sockaddr_dl" .
343 fields have the following values:
345 cmsg_len = CMSG_LEN(sizeof(struct sockaddr_dl))
346 cmsg_level = IPPROTO_IP
347 cmsg_type = IP_RECVIF
351 may be used to set the port range used for selecting a local port number
352 on a socket with an unspecified (zero) port number.
355 .Bl -tag -width IP_PORTRANGE_DEFAULT
356 .It Dv IP_PORTRANGE_DEFAULT
357 use the default range of values, normally
358 .Dv IPPORT_HIFIRSTAUTO
360 .Dv IPPORT_HILASTAUTO .
361 This is adjustable through the sysctl setting:
362 .Va net.inet.ip.portrange.first
364 .Va net.inet.ip.portrange.last .
365 .It Dv IP_PORTRANGE_HIGH
366 use a high range of values, normally
367 .Dv IPPORT_HIFIRSTAUTO
369 .Dv IPPORT_HILASTAUTO .
370 This is adjustable through the sysctl setting:
371 .Va net.inet.ip.portrange.hifirst
373 .Va net.inet.ip.portrange.hilast .
374 .It Dv IP_PORTRANGE_LOW
375 use a low range of ports, which are normally restricted to
376 privileged processes on
379 The range is normally from
382 .Li IPPORT_RESERVEDSTART
384 This is adjustable through the sysctl setting:
385 .Va net.inet.ip.portrange.lowfirst
387 .Va net.inet.ip.portrange.lowlast .
390 The range of privileged ports which only may be opened by
391 root-owned processes may be modified by the
392 .Va net.inet.ip.portrange.reservedlow
394 .Va net.inet.ip.portrange.reservedhigh
396 The values default to the traditional range,
400 (0 through 1023), respectively.
401 Note that these settings do not affect and are not accounted for in the
402 use or calculation of the other
403 .Va net.inet.ip.portrange
405 Changing these values departs from
407 tradition and has security
408 consequences that the administrator should carefully evaluate before
409 modifying these settings.
411 Ports are allocated at random within the specified port range in order
412 to increase the difficulty of random spoofing attacks.
413 In scenarios such as benchmarking, this behavior may be undesirable.
415 .Va net.inet.ip.portrange.randomized
416 can be used to toggle randomization off.
418 .Va net.inet.ip.portrange.randomcps
419 ports have been allocated in the last second, then return to sequential
421 Return to random allocation only once the current port allocation rate
423 .Va net.inet.ip.portrange.randomcps
425 .Va net.inet.ip.portrange.randomtime
427 The default values for
428 .Va net.inet.ip.portrange.randomcps
430 .Va net.inet.ip.portrange.randomtime
431 are 10 port allocations per second and 45 seconds correspondingly.
432 .Ss "Multicast Options"
434 multicasting is supported only on
440 and only on networks where the interface
441 driver supports multicasting.
445 option changes the time-to-live (TTL)
446 for outgoing multicast datagrams
447 in order to control the scope of the multicasts:
449 u_char ttl; /* range: 0 to 255, default = 1 */
450 setsockopt(s, IPPROTO_IP, IP_MULTICAST_TTL, &ttl, sizeof(ttl));
453 Datagrams with a TTL of 1 are not forwarded beyond the local network.
454 Multicast datagrams with a TTL of 0 will not be transmitted on any network,
455 but may be delivered locally if the sending host belongs to the destination
456 group and if multicast loopback has not been disabled on the sending socket
458 Multicast datagrams with TTL greater than 1 may be forwarded
459 to other networks if a multicast router is attached to the local network.
461 For hosts with multiple interfaces, where an interface has not
462 been specified for a multicast group membership,
463 each multicast transmission is sent from the primary network interface.
466 option overrides the default for
467 subsequent transmissions from a given socket:
470 setsockopt(s, IPPROTO_IP, IP_MULTICAST_IF, &addr, sizeof(addr));
473 where "addr" is the local
475 address of the desired interface or
477 to specify the default interface.
479 To specify an interface by index, an instance of
481 may be passed instead.
484 member should be set to the index of the desired interface,
485 or 0 to specify the default interface.
486 The kernel differentiates between these two structures by their size.
491 .Em not recommended ,
492 as multicast memberships are scoped to each
493 individual interface.
494 It is supported for legacy use only by applications,
495 such as routing daemons, which expect to
496 be able to transmit link-local IPv4 multicast datagrams (224.0.0.0/24)
497 on multiple interfaces,
498 without requesting an individual membership for each interface.
501 An interface's local IP address and multicast capability can
507 Normal applications should not need to use this option.
509 If a multicast datagram is sent to a group to which the sending host itself
510 belongs (on the outgoing interface), a copy of the datagram is, by default,
511 looped back by the IP layer for local delivery.
513 .Dv IP_MULTICAST_LOOP
514 option gives the sender explicit control
515 over whether or not subsequent datagrams are looped back:
517 u_char loop; /* 0 = disable, 1 = enable (default) */
518 setsockopt(s, IPPROTO_IP, IP_MULTICAST_LOOP, &loop, sizeof(loop));
522 improves performance for applications that may have no more than one
523 instance on a single host (such as a routing daemon), by eliminating
524 the overhead of receiving their own transmissions.
525 It should generally not
526 be used by applications for which there may be more than one instance on a
527 single host (such as a conferencing program) or for which the sender does
528 not belong to the destination group (such as a time querying program).
531 .Va net.inet.ip.mcast.loop
532 controls the default setting of the
533 .Dv IP_MULTICAST_LOOP
534 socket option for new sockets.
536 A multicast datagram sent with an initial TTL greater than 1 may be delivered
537 to the sending host on a different interface from that on which it was sent,
538 if the host belongs to the destination group on that other interface.
539 The loopback control option has no effect on such delivery.
541 A host must become a member of a multicast group before it can receive
542 datagrams sent to the group.
543 To join a multicast group, use the
544 .Dv IP_ADD_MEMBERSHIP
548 setsockopt(s, IPPROTO_IP, IP_ADD_MEMBERSHIP, &mreq, sizeof(mreq));
553 is the following structure:
556 struct in_addr imr_multiaddr; /* IP multicast address of group */
557 struct in_addr imr_interface; /* local IP address of interface */
564 address of a particular multicast-capable interface if
565 the host is multihomed.
568 to choose the default interface, although this is not recommended;
569 this is considered to be the first interface corresponding
570 to the default route.
571 Otherwise, the first multicast-capable interface
572 configured in the system will be used.
578 member is within the network range
580 it is treated as an interface index in the system interface MIB,
581 as per the RIP Version 2 MIB Extension (RFC-1724).
584 since 7.0, this behavior is no longer supported.
586 instead use the RFC 3678 multicast source filter APIs; in particular,
587 .Dv MCAST_JOIN_GROUP .
590 .Dv IP_MAX_MEMBERSHIPS
591 memberships may be added on a single socket.
592 Membership is associated with a single interface;
593 programs running on multihomed hosts may need to
594 join the same group on more than one interface.
596 To drop a membership, use:
599 setsockopt(s, IPPROTO_IP, IP_DROP_MEMBERSHIP, &mreq, sizeof(mreq));
604 contains the same values as used to add the membership.
605 Memberships are dropped when the socket is closed or the process exits.
606 .\" TODO: Update this piece when IPv4 source-address selection is implemented.
608 The IGMP protocol uses the primary IP address of the interface
609 as its identifier for group membership.
610 This is the first IP address configured on the interface.
611 If this address is removed or changed, the results are
612 undefined, as the IGMP membership state will then be inconsistent.
613 If multiple IP aliases are configured on the same interface,
614 they will be ignored.
616 This shortcoming was addressed in IPv6; MLDv2 requires
617 that the unique link-local address for an interface is
618 used to identify an MLDv2 listener.
619 .Ss "Source-Specific Multicast Options"
622 the use of Source-Specific Multicast (SSM) is supported.
623 These extensions require an IGMPv3 multicast router in order to
624 make best use of them.
625 If a legacy multicast router is present on the link,
627 will simply downgrade to the version of IGMP spoken by the router,
628 and the benefits of source filtering on the upstream link
629 will not be present, although the kernel will continue to
630 squelch transmissions from blocked sources.
632 Each group membership on a socket now has a filter mode:
633 .Bl -tag -width MCAST_EXCLUDE
635 Datagrams sent to this group are accepted,
636 unless the source is in a list of blocked source addresses.
638 Datagrams sent to this group are accepted
639 only if the source is in a list of accepted source addresses.
642 Groups joined using the legacy
643 .Dv IP_ADD_MEMBERSHIP
644 option are placed in exclusive-mode,
645 and are able to request that certain sources are blocked or allowed.
647 .Em delta-based API .
649 To block a multicast source on an existing group membership:
651 struct ip_mreq_source mreqs;
652 setsockopt(s, IPPROTO_IP, IP_BLOCK_SOURCE, &mreqs, sizeof(mreqs));
657 is the following structure:
659 struct ip_mreq_source {
660 struct in_addr imr_multiaddr; /* IP multicast address of group */
661 struct in_addr imr_sourceaddr; /* IP address of source */
662 struct in_addr imr_interface; /* local IP address of interface */
666 should be set to the address of the source to be blocked.
668 To unblock a multicast source on an existing group:
670 struct ip_mreq_source mreqs;
671 setsockopt(s, IPPROTO_IP, IP_UNBLOCK_SOURCE, &mreqs, sizeof(mreqs));
677 .Dv IP_UNBLOCK_SOURCE
680 for inclusive-mode group memberships.
682 To join a multicast group in
684 mode with a single source,
685 or add another source to an existing inclusive-mode membership:
687 struct ip_mreq_source mreqs;
688 setsockopt(s, IPPROTO_IP, IP_ADD_SOURCE_MEMBERSHIP, &mreqs, sizeof(mreqs));
691 To leave a single source from an existing group in inclusive mode:
693 struct ip_mreq_source mreqs;
694 setsockopt(s, IPPROTO_IP, IP_DROP_SOURCE_MEMBERSHIP, &mreqs, sizeof(mreqs));
696 If this is the last accepted source for the group, the membership
700 .Dv IP_ADD_SOURCE_MEMBERSHIP
702 .Dv IP_DROP_SOURCE_MEMBERSHIP
705 for exclusive-mode group memberships.
706 However, both exclusive and inclusive mode memberships
707 support the use of the
709 documented in RFC 3678.
710 For management of source filter lists using this API,
715 .Va net.inet.ip.mcast.maxsocksrc
717 .Va net.inet.ip.mcast.maxgrpsrc
718 are used to specify an upper limit on the number of per-socket and per-group
719 source filter entries which the kernel may allocate.
720 .\"-----------------------
724 sockets are connectionless,
725 and are normally used with the
731 call may also be used to fix the destination for future
732 packets (in which case the
740 system calls may be used).
744 is 0, the default protocol
747 packets, and only incoming packets destined for that protocol
751 is non-zero, that protocol number will be used on outgoing packets
752 and to filter incoming packets.
754 Outgoing packets automatically have an
757 them (based on the destination address and the protocol
758 number the socket is created with),
762 Incoming packets are received with
764 header and options intact, except for
768 fields converted to host byte order.
771 indicates the complete IP header is included with the data
772 and may be used only with the
776 #include <netinet/in_systm.h>
777 #include <netinet/ip.h>
779 int hincl = 1; /* 1 = on, 0 = off */
780 setsockopt(s, IPPROTO_IP, IP_HDRINCL, &hincl, sizeof(hincl));
785 releases, the program must set all
786 the fields of the IP header, including the following:
788 ip->ip_v = IPVERSION;
789 ip->ip_hl = hlen >> 2;
790 ip->ip_id = 0; /* 0 means kernel set appropriate value */
800 be provided in host byte order.
801 All other fields must be provided in network byte order.
804 for more information on network byte order.
807 field is set to 0 then the kernel will choose an
809 If the header source address is set to
811 the kernel will choose an appropriate address.
813 A socket operation may fail with one of the following errors returned:
816 when trying to establish a connection on a socket which
817 already has one, or when trying to send a datagram with the destination
818 address specified and the socket is already connected;
820 when trying to send a datagram, but
821 no destination address is specified, and the socket has not been
824 when the system runs out of memory for
825 an internal data structure;
826 .It Bq Er EADDRNOTAVAIL
827 when an attempt is made to create a
828 socket with a network address for which no network interface
831 when an attempt is made to create
832 a raw IP socket by a non-privileged process.
835 The following errors specific to
837 may occur when setting or getting
842 An unknown socket option name was given.
844 The IP option field was improperly formed;
845 an option field was shorter than the minimum value
846 or longer than the option buffer provided.
849 The following errors may occur when attempting to send
860 field was not equal to the length of the datagram written to the socket.
877 .%T "Socket Interface Extensions for Multicast Source Filters"
888 structure appeared in
893 packets received on raw IP sockets had the