xref: /openbsd/share/man/man4/ip6.4 (revision 3d8817e4)
1.\"	$OpenBSD: ip6.4,v 1.23 2009/08/31 04:46:32 jmc Exp $
2.\"
3.\" Copyright (c) 1983, 1991, 1993
4.\"	The Regents of the University of California.  All rights reserved.
5.\"
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\" 1. Redistributions of source code must retain the above copyright
10.\"    notice, this list of conditions and the following disclaimer.
11.\" 2. Redistributions in binary form must reproduce the above copyright
12.\"    notice, this list of conditions and the following disclaimer in the
13.\"    documentation and/or other materials provided with the distribution.
14.\" 3. Neither the name of the University nor the names of its contributors
15.\"    may be used to endorse or promote products derived from this software
16.\"    without specific prior written permission.
17.\"
18.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
19.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
20.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
21.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
22.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
23.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
24.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
25.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
26.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
27.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28.\" SUCH DAMAGE.
29.Dd $Mdocdate: August 31 2009 $
30.Dt IP6 4
31.Os
32.Sh NAME
33.Nm ip6
34.Nd Internet Protocol version 6 (IPv6) network layer
35.Sh SYNOPSIS
36.In sys/socket.h
37.In netinet/in.h
38.Ft int
39.Fn socket AF_INET6 SOCK_RAW proto
40.Sh DESCRIPTION
41The IPv6 network layer is used by the IPv6 protocol family for
42transporting data.
43IPv6 packets contain an IPv6 header that is not provided as part of the
44payload contents when passed to an application.
45IPv6 header options affect the behavior of this protocol and may be used
46by high-level protocols (such as the
47.Xr tcp 4
48and
49.Xr udp 4
50protocols) as well as directly by
51.Dq raw sockets ,
52which process IPv6 messages at a lower-level and may be useful for
53developing new protocols and special-purpose applications.
54.Ss Header
55All IPv6 packets begin with an IPv6 header.
56When data received by the kernel are passed to the application, this
57header is not included in buffer, even when raw sockets are being used.
58Likewise, when data are sent to the kernel for transmit from the
59application, the buffer is not examined for an IPv6 header:
60the kernel always constructs the header.
61To directly access IPv6 headers from received packets and specify them
62as part of the buffer passed to the kernel, link-level access
63.Po
64.Xr bpf 4 ,
65for example
66.Pc
67must instead be utilized.
68.Pp
69The header has the following definition:
70.Bd -literal -offset indent
71struct ip6_hdr {
72     union {
73          struct ip6_hdrctl {
74               u_int32_t ip6_un1_flow;	/* 20 bits of flow ID */
75               u_int16_t ip6_un1_plen;	/* payload length */
76               u_int8_t	 ip6_un1_nxt;	/* next header */
77               u_int8_t	 ip6_un1_hlim;	/* hop limit */
78          } ip6_un1;
79          u_int8_t ip6_un2_vfc;   /* version and class */
80     } ip6_ctlun;
81     struct in6_addr ip6_src;	/* source address */
82     struct in6_addr ip6_dst;	/* destination address */
83} __packed;
84
85#define ip6_vfc		ip6_ctlun.ip6_un2_vfc
86#define ip6_flow	ip6_ctlun.ip6_un1.ip6_un1_flow
87#define ip6_plen	ip6_ctlun.ip6_un1.ip6_un1_plen
88#define ip6_nxt		ip6_ctlun.ip6_un1.ip6_un1_nxt
89#define ip6_hlim	ip6_ctlun.ip6_un1.ip6_un1_hlim
90#define ip6_hops	ip6_ctlun.ip6_un1.ip6_un1_hlim
91.Ed
92.Pp
93All fields are in network-byte order.
94Any options specified (see
95.Sx Options
96below) must also be specified in network-byte order.
97.Pp
98.Va ip6_flow
99specifies the flow ID.
100.Va ip6_plen
101specifies the payload length.
102.Va ip6_nxt
103specifies the type of the next header.
104.Va ip6_hlim
105specifies the hop limit.
106.Pp
107The top 4 bits of
108.Va ip6_vfc
109specify the class and the bottom 4 bits specify the version.
110.Pp
111.Va ip6_src
112and
113.Va ip6_dst
114specify the source and destination addresses.
115.Pp
116The IPv6 header may be followed by any number of extension headers that start
117with the following generic definition:
118.Bd -literal -offset indent
119struct ip6_ext {
120     u_int8_t ip6e_nxt;
121     u_int8_t ip6e_len;
122} __packed;
123.Ed
124.Ss Options
125IPv6 allows header options on packets to manipulate the behavior of the
126protocol.
127These options and other control requests are accessed with the
128.Xr getsockopt 2
129and
130.Xr setsockopt 2
131system calls at level
132.Dv IPPROTO_IPV6
133and by using ancillary data in
134.Xr recvmsg 2
135and
136.Xr sendmsg 2 .
137They can be used to access most of the fields in the IPv6 header and
138extension headers.
139.Pp
140The following socket options are supported:
141.Bl -tag -width Ds
142.\" .It Dv IPV6_OPTIONS
143.It Dv IPV6_UNICAST_HOPS Fa "int *"
144Get or set the default hop limit header field for outgoing unicast
145datagrams sent on this socket.
146A value of \-1 resets to the default value.
147.\" .It Dv IPV6_RECVOPTS Fa "int *"
148.\" Get or set the status of whether all header options will be
149.\" delivered along with the datagram when it is received.
150.\" .It Dv IPV6_RECVRETOPTS Fa "int *"
151.\" Get or set the status of whether header options will be delivered
152.\" for reply.
153.\" .It Dv IPV6_RECVDSTADDR Fa "int *"
154.\" Get or set the status of whether datagrams are received with
155.\" destination addresses.
156.\" .It Dv IPV6_RETOPTS
157.\" Get or set IPv6 options.
158.It Dv IPV6_MULTICAST_IF Fa "u_int *"
159Get or set the interface from which multicast packets will be sent.
160For hosts with multiple interfaces, each multicast transmission is sent
161from the primary network interface.
162The interface is specified as its index as provided by
163.Xr if_nametoindex 3 .
164A value of zero specifies the default interface.
165.It Dv IPV6_MULTICAST_HOPS Fa "int *"
166Get or set the default hop limit header field for outgoing multicast
167datagrams sent on this socket.
168This option controls the scope of multicast datagram transmissions.
169.Pp
170Datagrams with a hop limit of 1 are not forwarded beyond the local
171network.
172Multicast datagrams with a hop limit of zero will not be transmitted on
173any network but may be delivered locally if the sending host belongs to
174the destination group and if multicast loopback (see below) has not been
175disabled on the sending socket.
176Multicast datagrams with a hop limit greater than 1 may be forwarded to
177the other networks if a multicast router (such as
178.Xr mrouted 8 )
179is attached to the local network.
180.It Dv IPV6_MULTICAST_LOOP Fa "u_int *"
181Get or set the status of whether multicast datagrams will be looped back
182for local delivery when a multicast datagram is sent to a group to which
183the sending host belongs.
184.Pp
185This option improves performance for applications that may have no more
186than one instance on a single host (such as a router daemon) by
187eliminating the overhead of receiving their own transmissions.
188It should generally not be used by applications for which there may be
189more than one instance on a single host (such as a conferencing program)
190or for which the sender does not belong to the destination group
191(such as a time-querying program).
192.Pp
193A multicast datagram sent with an initial hop limit greater than 1 may
194be delivered to the sending host on a different interface from that on
195which it was sent if the host belongs to the destination group on that
196other interface.
197The multicast loopback control option has no effect on such delivery.
198.It Dv IPV6_JOIN_GROUP Fa "struct ipv6_mreq *"
199Join a multicast group.
200A host must become a member of a multicast group before it can receive
201datagrams sent to the group.
202.Bd -literal
203struct ipv6_mreq {
204	struct in6_addr	ipv6mr_multiaddr;
205	unsigned int	ipv6mr_interface;
206};
207.Ed
208.Pp
209.Va ipv6mr_interface
210may be set to zeroes to choose the default multicast interface or to the
211index of a particular multicast-capable interface if the host is
212multihomed.
213Membership is associated with a single interface; programs running on
214multihomed hosts may need to join the same group on more than one
215interface.
216.Pp
217If the multicast address is unspecified (i.e., all zeroes), messages
218from all multicast addresses will be accepted by this group.
219Note that setting to this value requires superuser privileges.
220.It Dv IPV6_LEAVE_GROUP Fa "struct ipv6_mreq *"
221Drop membership from the associated multicast group.
222Memberships are automatically dropped when the socket is closed or when
223the process exits.
224.It Dv IPV6_PORTRANGE Fa "int *"
225Get or set the allocation policy of ephemeral ports for when the kernel
226automatically binds a local address to this socket.
227The following values are available:
228.Pp
229.Bl -tag -width IPV6_PORTRANGE_DEFAULT -compact
230.It Dv IPV6_PORTRANGE_DEFAULT
231Use the regular range of non-reserved ports (varies, see
232.Xr sysctl 8 ) .
233.It Dv IPV6_PORTRANGE_HIGH
234Use a high range (varies, see
235.Xr sysctl 8 ) .
236.It Dv IPV6_PORTRANGE_LOW
237Use a low, reserved range (600\-1023).
238.El
239.It Dv IPV6_PKTINFO Fa "int *"
240Get or set whether additional information about subsequent packets will
241be provided as ancillary data along with the payload in subsequent
242.Xr recvmsg 2
243calls.
244The information is stored in the following structure in the ancillary
245data returned:
246.Bd -literal
247struct in6_pktinfo {
248	struct in6_addr ipi6_addr;    /* src/dst IPv6 address */
249	unsigned int    ipi6_ifindex; /* send/recv if index */
250};
251.Ed
252.It Dv IPV6_HOPLIMIT Fa "int *"
253Get or set whether the hop limit header field from subsequent packets
254will be provided as ancillary data along with the payload in subsequent
255.Xr recvmsg 2
256calls.
257The value is stored as an
258.Vt int
259in the ancillary data returned.
260.\" .It Dv IPV6_NEXTHOP Fa "int *"
261.\" Get or set whether the address of the next hop for subsequent
262.\" packets will be provided as ancillary data along with the payload in
263.\" subsequent
264.\" .Xr recvmsg 2
265.\" calls.
266.\" The option is stored as a
267.\" .Vt sockaddr
268.\" structure in the ancillary data returned.
269.\" .Pp
270.\" This option requires superuser privileges.
271.It Dv IPV6_HOPOPTS Fa "int *"
272Get or set whether the hop-by-hop options from subsequent packets will be
273provided as ancillary data along with the payload in subsequent
274.Xr recvmsg 2
275calls.
276The option is stored in the following structure in the ancillary data
277returned:
278.Bd -literal
279struct ip6_hbh {
280	u_int8_t ip6h_nxt;	/* next header */
281	u_int8_t ip6h_len;	/* length in units of 8 octets */
282/* followed by options */
283} __packed;
284.Ed
285.Pp
286The
287.Fn inet6_option_space
288routine and family of routines may be used to manipulate this data.
289.Pp
290This option requires superuser privileges.
291.It Dv IPV6_DSTOPTS Fa "int *"
292Get or set whether the destination options from subsequent packets will
293be provided as ancillary data along with the payload in subsequent
294.Xr recvmsg 2
295calls.
296The option is stored in the following structure in the ancillary data
297returned:
298.Bd -literal
299struct ip6_dest {
300	u_int8_t ip6d_nxt;	/* next header */
301	u_int8_t ip6d_len;	/* length in units of 8 octets */
302/* followed by options */
303} __packed;
304.Ed
305.Pp
306The
307.Fn inet6_option_space
308routine and family of routines may be used to manipulate this data.
309.Pp
310This option requires superuser privileges.
311.It Dv IPV6_RTHDR Fa "int *"
312Get or set whether the routing header from subsequent packets will be
313provided as ancillary data along with the payload in subsequent
314.Xr recvmsg 2
315calls.
316The header is stored in the following structure in the ancillary data
317returned:
318.Bd -literal
319struct ip6_rthdr {
320	u_int8_t ip6r_nxt;	/* next header */
321	u_int8_t ip6r_len;	/* length in units of 8 octets */
322	u_int8_t ip6r_type;	/* routing type */
323	u_int8_t ip6r_segleft;	/* segments left */
324/* followed by routing-type-specific data */
325} __packed;
326.Ed
327.Pp
328The
329.Fn inet6_option_space
330routine and family of routines may be used to manipulate this data.
331.Pp
332This option requires superuser privileges.
333.It Dv IPV6_PKTOPTIONS Fa "struct cmsghdr *"
334Get or set all header options and extension headers at one time on the
335last packet sent or received on the socket.
336All options must fit within the size of an mbuf (see
337.Xr mbuf 9 ) .
338Options are specified as a series of
339.Vt cmsghdr
340structures followed by corresponding values.
341.Va cmsg_level
342is set to
343.Dv IPPROTO_IPV6 ,
344.Va cmsg_type
345to one of the other values in this list, and trailing data to the option
346value.
347When setting options, if the length
348.Va optlen
349to
350.Xr setsockopt 2
351is zero, all header options will be reset to their default values.
352Otherwise, the length should specify the size the series of control
353messages consumes.
354.Pp
355Instead of using
356.Xr sendmsg 2
357to specify option values, the ancillary data used in these calls that
358correspond to the desired header options may be directly specified as
359the control message in the series of control messages provided as the
360argument to
361.Xr setsockopt 2 .
362.It Dv IPV6_CHECKSUM Fa "int *"
363Get or set the byte offset into a packet where the 16-bit checksum is
364located.
365When set, this byte offset is where incoming packets will be expected
366to have checksums of their data stored and where outgoing packets will
367have checksums of their data computed and stored by the kernel.
368A value of \-1 specifies that no checksums will be checked on incoming
369packets and that no checksums will be computed or stored on outgoing
370packets.
371The offset of the checksum for ICMPv6 sockets cannot be relocated or
372turned off.
373.It Dv IPV6_V6ONLY Fa "int *"
374Get or set whether only IPv6 connections can be made to this socket.
375For wildcard sockets, this can restrict connections to IPv6 only.
376With
377.Ox
378IPv6 sockets are always IPv6-only, so the socket option is read-only
379(not modifiable).
380.It Dv IPV6_FAITH Fa "int *"
381Get or set the status of whether
382.Xr faith 4
383connections can be made to this socket.
384.It Dv IPV6_USE_MIN_MTU Fa "int *"
385Get or set whether the minimal IPv6 maximum transmission unit (MTU) size
386will be used to avoid fragmentation from occurring for subsequent
387outgoing datagrams.
388.It Dv IPV6_AUTH_LEVEL Fa "int *"
389Get or set the
390.Xr ipsec 4
391authentication level.
392.It Dv IPV6_ESP_TRANS_LEVEL Fa "int *"
393Get or set the ESP transport level.
394.It Dv IPV6_ESP_NETWORK_LEVEL Fa "int *"
395Get or set the ESP encapsulation level.
396.It Dv IPV6_IPCOMP_LEVEL Fa "int *"
397Get or set the
398.Xr ipcomp 4
399level.
400.El
401.Pp
402The
403.Dv IPV6_PKTINFO ,
404.\" .Dv IPV6_NEXTHOP ,
405.Dv IPV6_HOPLIMIT ,
406.Dv IPV6_HOPOPTS ,
407.Dv IPV6_DSTOPTS ,
408and
409.Dv IPV6_RTHDR
410options will return ancillary data along with payload contents in subsequent
411.Xr recvmsg 2
412calls with
413.Va cmsg_level
414set to
415.Dv IPPROTO_IPV6
416and
417.Va cmsg_type
418set to respective option name value (e.g.,
419.Dv IPV6_HOPTLIMIT ) .
420These options may also be used directly as ancillary
421.Va cmsg_type
422values in
423.Xr sendmsg 2
424to set options on the packet being transmitted by the call.
425The
426.Va cmsg_level
427value must be
428.Dv IPPROTO_IPV6 .
429For these options, the ancillary data object value format is the same
430as the value returned as explained for each when received with
431.Xr recvmsg 2 .
432.Pp
433Note that using
434.Xr sendmsg 2
435to specify options on particular packets works only on UDP and raw sockets.
436To manipulate header options for packets on TCP sockets, only the socket
437options may be used.
438.Pp
439In some cases, there are multiple APIs defined for manipulating an IPv6
440header field.
441A good example is the outgoing interface for multicast datagrams, which
442can be set by the
443.Dv IPV6_MULTICAST_IF
444socket option, through the
445.Dv IPV6_PKTINFO
446option, and through the
447.Va sin6_scope_id
448field of the socket address passed to the
449.Xr sendto 2
450system call.
451.Pp
452Resolving these conflicts is implementation dependent.
453This implementation determines the value in the following way:
454options specified by using ancillary data (i.e.,
455.Xr sendmsg 2 )
456are considered first,
457options specified by using
458.Dv IPV6_PKTOPTIONS
459to set
460.Dq sticky
461options are considered second,
462options specified by using the individual, basic, and direct socket
463options (e.g.,
464.Dv IPV6_UNICAST_HOPS )
465are considered third,
466and options specified in the socket address supplied to
467.Xr sendto 2
468are the last choice.
469.Ss Multicasting
470IPv6 multicasting is supported only on
471.Dv AF_INET6
472sockets of type
473.Dv SOCK_DGRAM
474and
475.Dv SOCK_RAW ,
476and only on networks where the interface driver supports
477multicasting.
478Socket options (see above) that manipulate membership of
479multicast groups and other multicast options include
480.Dv IPV6_MULTICAST_IF ,
481.Dv IPV6_MULTICAST_HOPS ,
482.Dv IPV6_MULTICAST_LOOP ,
483.Dv IPV6_LEAVE_GROUP ,
484and
485.Dv IPV6_JOIN_GROUP .
486.Ss Raw Sockets
487Raw IPv6 sockets are connectionless and are normally used with the
488.Xr sendto 2
489and
490.Xr recvfrom 2
491calls, although the
492.Xr connect 2
493call may be used to fix the destination address for future outgoing
494packets so that
495.Xr send 2
496may instead be used and the
497.Xr bind 2
498call may be used to fix the source address for future outgoing
499packets instead of having the kernel choose a source address.
500.Pp
501By using
502.Xr connect 2
503or
504.Xr bind 2 ,
505raw socket input is constrained to only packets with their
506source address matching the socket destination address if
507.Xr connect 2
508was used and to packets with their destination address
509matching the socket source address if
510.Xr bind 2
511was used.
512.Pp
513If the
514.Ar proto
515argument to
516.Xr socket 2
517is zero, the default protocol
518.Pq Dv IPPROTO_RAW
519is used for outgoing packets.
520For incoming packets, protocols recognized by kernel are
521.Sy not
522passed to the application socket (e.g.,
523.Xr tcp 4
524and
525.Xr udp 4 )
526except for some ICMPv6 messages.
527The ICMPv6 messages not passed to raw sockets include echo, timestamp,
528and address mask requests.
529If
530.Ar proto
531is non-zero, only packets with this protocol will be passed to the
532socket.
533.Pp
534IPv6 fragments are also not passed to application sockets until
535they have been reassembled.
536If reception of all packets is desired, link-level access (such as
537.Xr bpf 4 )
538must be used instead.
539.Pp
540Outgoing packets automatically have an IPv6 header prepended to them
541(based on the destination address and the protocol number the socket
542was created with).
543Incoming packets are received by an application without the IPv6 header
544or any extension headers.
545.Pp
546Outgoing packets will be fragmented automatically by the kernel if they
547are too large.
548Incoming packets will be reassembled before being sent to the raw socket,
549so packet fragments or fragment headers will never be seen on a raw socket.
550.Sh EXAMPLES
551The following determines the hop limit on the next packet received:
552.Bd -literal
553struct iovec iov[2];
554u_char buf[BUFSIZ];
555struct cmsghdr *cm;
556struct msghdr m;
557int found, optval;
558u_char data[2048];
559
560/* Create socket. */
561
562(void)memset(&m, 0, sizeof(m));
563(void)memset(&iov, 0, sizeof(iov));
564
565iov[0].iov_base = data;		/* buffer for packet payload */
566iov[0].iov_len = sizeof(data);	/* expected packet length */
567
568m.msg_name = &from;		/* sockaddr_in6 of peer */
569m.msg_namelen = sizeof(from);
570m.msg_iov = iov;
571m.msg_iovlen = 1;
572m.msg_control = (caddr_t)buf;	/* buffer for control messages */
573m.msg_controllen = sizeof(buf);
574
575/*
576 * Enable the hop limit value from received packets to be
577 * returned along with the payload.
578 */
579optval = 1;
580if (setsockopt(s, IPPROTO_IPV6, IPV6_HOPLIMIT, &optval,
581    sizeof(optval)) == -1)
582	err(1, "setsockopt");
583
584found = 0;
585while (!found) {
586	if (recvmsg(s, &m, 0) == -1)
587		err(1, "recvmsg");
588	for (cm = CMSG_FIRSTHDR(&m); cm != NULL;
589	     cm = CMSG_NXTHDR(&m, cm)) {
590		if (cm->cmsg_level == IPPROTO_IPV6 &&
591		    cm->cmsg_type == IPV6_HOPLIMIT &&
592		    cm->cmsg_len == CMSG_LEN(sizeof(int))) {
593			found = 1;
594			(void)printf("hop limit: %d\en",
595			    *(int *)CMSG_DATA(cm));
596			break;
597		}
598	}
599}
600.Ed
601.Sh DIAGNOSTICS
602A socket operation may fail with one of the following errors returned:
603.Bl -tag -width EADDRNOTAVAILxx
604.It Bq Er EISCONN
605when trying to establish a connection on a socket which
606already has one or when trying to send a datagram with the destination
607address specified and the socket is already connected.
608.It Bq Er ENOTCONN
609when trying to send a datagram, but
610no destination address is specified, and the socket hasn't been
611connected.
612.It Bq Er ENOBUFS
613when the system runs out of memory for
614an internal data structure.
615.It Bq Er EADDRNOTAVAIL
616when an attempt is made to create a
617socket with a network address for which no network interface
618exists.
619.It Bq Er EACCES
620when an attempt is made to create
621a raw IPv6 socket by a non-privileged process.
622.El
623.Pp
624The following errors specific to IPv6 may occur when setting or getting
625header options:
626.Bl -tag -width EADDRNOTAVAILxx
627.It Bq Er EINVAL
628An unknown socket option name was given.
629.It Bq Er EINVAL
630An ancillary data object was improperly formed.
631.El
632.Sh SEE ALSO
633.Xr getsockopt 2 ,
634.Xr recv 2 ,
635.Xr send 2 ,
636.Xr setsockopt 2 ,
637.Xr socket 2 ,
638.\" .Xr inet6_option_space 3 ,
639.\" .Xr inet6_rthdr_space 3 ,
640.Xr if_nametoindex 3 ,
641.Xr bpf 4 ,
642.Xr icmp6 4 ,
643.Xr inet6 4 ,
644.Xr netintro 4 ,
645.Xr tcp 4 ,
646.Xr udp 4
647.Rs
648.%A W. Stevens
649.%A M. Thomas
650.%T Advanced Sockets API for IPv6
651.%R RFC 2292
652.%D February 1998
653.Re
654.Rs
655.%A S. Deering
656.%A R. Hinden
657.%T Internet Protocol, Version 6 (IPv6) Specification
658.%R RFC 2460
659.%D December 1998
660.Re
661.Rs
662.%A R. Gilligan
663.%A S. Thomson
664.%A J. Bound
665.%A W. Stevens
666.%T Basic Socket Interface Extensions for IPv6
667.%R RFC 2553
668.%D March 1999
669.Re
670.Rs
671.%A W. Stevens
672.%A B. Fenner
673.%A A. Rudoff
674.%T UNIX Network Programming, third edition
675.Re
676.Sh STANDARDS
677Most of the socket options are defined in RFC 2292 or RFC 2553.
678The
679.Dv IPV6_V6ONLY
680socket option is defined in RFC 3493.
681The
682.Dv IPV6_PORTRANGE
683socket option and the conflict resolution rule are not defined in the
684RFCs and should be considered implementation dependent.
685