1.\" $NetBSD: ip6.4,v 1.35 2024/09/04 02:36:56 rin Exp $ 2.\" $KAME: ip6.4,v 1.23 2005/01/11 05:56:25 itojun Exp $ 3.\" $OpenBSD: ip6.4,v 1.21 2005/01/06 03:50:46 itojun Exp $ 4.\" 5.\" Copyright (c) 1983, 1991, 1993 6.\" The Regents of the University of California. All rights reserved. 7.\" 8.\" Redistribution and use in source and binary forms, with or without 9.\" modification, are permitted provided that the following conditions 10.\" are met: 11.\" 1. Redistributions of source code must retain the above copyright 12.\" notice, this list of conditions and the following disclaimer. 13.\" 2. Redistributions in binary form must reproduce the above copyright 14.\" notice, this list of conditions and the following disclaimer in the 15.\" documentation and/or other materials provided with the distribution. 16.\" 3. 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. 19.\" 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 30.\" SUCH DAMAGE. 31.Dd September 4, 2024 32.Dt IP6 4 33.Os 34.Sh NAME 35.Nm ip6 36.Nd Internet Protocol version 6 (IPv6) network layer 37.Sh SYNOPSIS 38.In sys/socket.h 39.In netinet/in.h 40.Ft int 41.Fn socket AF_INET6 SOCK_RAW proto 42.Sh DESCRIPTION 43The IPv6 network layer is used by the IPv6 protocol family for 44transporting data. 45IPv6 packets contain an IPv6 header that is not provided as part of the 46payload contents when passed to an application. 47IPv6 header options affect the behavior of this protocol and may be used 48by high-level protocols (such as the 49.Xr tcp 4 50and 51.Xr udp 4 52protocols) as well as directly by 53.Dq raw sockets , 54which process IPv6 messages at a lower-level and may be useful for 55developing new protocols and special-purpose applications. 56.Ss Header 57All IPv6 packets begin with an IPv6 header. 58When data received by the kernel are passed to the application, this 59header is not included in buffer, even when raw sockets are being used. 60Likewise, when data are sent to the kernel for transmit from the 61application, the buffer is not examined for an IPv6 header: 62the kernel always constructs the header. 63To directly access IPv6 headers from received packets and specify them 64as part of the buffer passed to the kernel, link-level access 65.Po 66.Xr bpf 4 , 67for example 68.Pc 69must be used instead. 70.Pp 71The header has the following definition: 72.Bd -literal -offset indent 73struct ip6_hdr { 74 union { 75 struct ip6_hdrctl { 76 uint32_t ip6_un1_flow; /* 20 bits of flow ID */ 77 uint16_t ip6_un1_plen; /* payload length */ 78 uint8_t ip6_un1_nxt; /* next header */ 79 uint8_t ip6_un1_hlim; /* hop limit */ 80 } ip6_un1; 81 uint8_t ip6_un2_vfc; /* version and class */ 82 } ip6_ctlun; 83 struct in6_addr ip6_src; /* source address */ 84 struct in6_addr ip6_dst; /* destination address */ 85} __packed; 86 87#define ip6_vfc ip6_ctlun.ip6_un2_vfc 88#define ip6_flow ip6_ctlun.ip6_un1.ip6_un1_flow 89#define ip6_plen ip6_ctlun.ip6_un1.ip6_un1_plen 90#define ip6_nxt ip6_ctlun.ip6_un1.ip6_un1_nxt 91#define ip6_hlim ip6_ctlun.ip6_un1.ip6_un1_hlim 92#define ip6_hops ip6_ctlun.ip6_un1.ip6_un1_hlim 93.Ed 94.Pp 95All fields are in network-byte order. 96Any options specified (see 97.Sx Options 98below) must also be specified in network-byte order. 99.Pp 100.Va ip6_flow 101specifies the flow ID. 102.Va ip6_plen 103specifies the payload length. 104.Va ip6_nxt 105specifies the type of the next header. 106.Va ip6_hlim 107specifies the hop limit. 108.Pp 109The top 4 bits of 110.Va ip6_vfc 111specify the class and the bottom 4 bits specify the version. 112.Pp 113.Va ip6_src 114and 115.Va ip6_dst 116specify the source and destination addresses. 117.Pp 118The IPv6 header may be followed by any number of extension headers that start 119with the following generic definition: 120.Bd -literal -offset indent 121struct ip6_ext { 122 uint8_t ip6e_nxt; 123 uint8_t ip6e_len; 124} __packed; 125.Ed 126.Ss Options 127IPv6 allows header options on packets to manipulate the behavior of the 128protocol. 129These options and other control requests are accessed with the 130.Xr getsockopt 2 131and 132.Xr setsockopt 2 133system calls at level 134.Dv IPPROTO_IPV6 135and by using ancillary data in 136.Xr recvmsg 2 137and 138.Xr sendmsg 2 . 139They can be used to access most of the fields in the IPv6 header and 140extension headers. 141.Pp 142The following socket options are supported: 143.Bl -tag -width Ds 144.\" .It Dv IPV6_OPTIONS 145.It Dv IPV6_UNICAST_HOPS Fa "int *" 146Get or set the default hop limit header field for outgoing unicast 147datagrams sent on this socket. 148A value of \-1 resets to the default value. 149.\" .It Dv IPV6_RECVOPTS Fa "int *" 150.\" Get or set the status of whether all header options will be 151.\" delivered along with the datagram when it is received. 152.\" .It Dv IPV6_RECVRETOPTS Fa "int *" 153.\" Get or set the status of whether header options will be delivered 154.\" for reply. 155.\" .It Dv IPV6_RECVDSTADDR Fa "int *" 156.\" Get or set the status of whether datagrams are received with 157.\" destination addresses. 158.\" .It Dv IPV6_RETOPTS 159.\" Get or set IPv6 options. 160.It Dv IPV6_MULTICAST_IF Fa "u_int *" 161Get or set the interface from which multicast packets will be sent. 162For hosts with multiple interfaces, each multicast transmission is sent 163from the primary network interface. 164The interface is specified as its index as provided by 165.Xr if_nametoindex 3 . 166A value of zero specifies the default interface. 167.It Dv IPV6_MULTICAST_HOPS Fa "int *" 168Get or set the default hop limit header field for outgoing multicast 169datagrams sent on this socket. 170This option controls the scope of multicast datagram transmissions. 171.Pp 172Datagrams with a hop limit of 1 are not forwarded beyond the local 173network. 174Multicast datagrams with a hop limit of zero will not be transmitted on 175any network but may be delivered locally if the sending host belongs to 176the destination group and if multicast loopback (see below) has not been 177disabled on the sending socket. 178Multicast datagrams with a hop limit greater than 1 may be forwarded to 179the other networks if a multicast router (such as 180.Xr mrouted 8 ) 181is attached to the local network. 182.It Dv IPV6_MULTICAST_LOOP Fa "u_int *" 183Get or set the status of whether multicast datagrams will be looped back 184for local delivery when a multicast datagram is sent to a group to which 185the sending host belongs. 186.Pp 187This option improves performance for applications that may have no more 188than one instance on a single host (such as a router daemon) by 189eliminating the overhead of receiving their own transmissions. 190It should generally not be used by applications for which there may be 191more than one instance on a single host (such as a conferencing program) 192or for which the sender does not belong to the destination group 193(such as a time-querying program). 194.Pp 195A multicast datagram sent with an initial hop limit greater than 1 may 196be delivered to the sending host on a different interface from that on 197which it was sent if the host belongs to the destination group on that 198other interface. 199The multicast loopback control option has no effect on such delivery. 200.It Dv IPV6_JOIN_GROUP Fa "struct ipv6_mreq *" 201Join a multicast group. 202A host must become a member of a multicast group before it can receive 203datagrams sent to the group. 204.Bd -literal 205struct ipv6_mreq { 206 struct in6_addr ipv6mr_multiaddr; 207 unsigned int ipv6mr_interface; 208}; 209.Ed 210.Pp 211.Va ipv6mr_interface 212may be set to zeroes to choose the default multicast interface or to the 213index of a particular multicast-capable interface if the host is 214multihomed. 215Membership is associated with a single interface; programs running on 216multihomed hosts may need to join the same group on more than one 217interface. 218.Pp 219If the multicast address is unspecified (i.e., all zeroes), messages 220from all multicast addresses will be accepted by this group. 221Note that setting to this value requires superuser privileges. 222.It Dv IPV6_LEAVE_GROUP Fa "struct ipv6_mreq *" 223Drop membership from the associated multicast group. 224Memberships are automatically dropped when the socket is closed or when 225the process exits. 226.It Dv IPV6_IPSEC_POLICY Fa "struct sadb_x_policy *" 227Get or set IPSec policy for sockets. 228For example, 229.Bd -literal 230const char *policy = "in ipsec ah/transport//require"; 231char *buf = ipsec_set_policy(policy, strlen(policy)); 232setsockopt(s, IPPROTO_IPV6, IPV6_IPSEC_POLICY, buf, ipsec_get_policylen(buf)); 233.Ed 234.It Dv IPV6_PORTALGO Fa "int *" 235The 236.Dv IP_PORTALGO 237can be used to randomize the port selection. 238Valid algorithms are described in 239.Xr rfc6056 7 240and their respective constants are in 241.In netinet/portalgo.h . 242For example, 243.Bd -literal 244int algo = PORTALGO_ALGO_RANDOM_PICK; /* see <netinet/portalgo.h> */ 245setsockopt(s, IPPROTO_IPV6, IPV6_PORTALGO, &algo, sizeof(algo)); 246.Ed 247.Pp 248The port selection can be also viewed and controlled at a global level for all 249.Tn IPV6 250sockets using the following 251.Xr sysctl 7 252variables: 253.Dv net.inet6.ip6.anonportalgo.available 254and 255.Dv net.inet6.ip6.anonportalgo.selected . 256.It Dv IPV6_PORTRANGE Fa "int *" 257Get or set the allocation policy of ephemeral ports for when the kernel 258automatically binds a local address to this socket. 259The following values are available: 260.Pp 261.Bl -tag -width IPV6_PORTRANGE_DEFAULT -compact 262.It Dv IPV6_PORTRANGE_DEFAULT 263Use the regular range of non-reserved ports (varies, see 264.Xr sysctl 8 ) . 265.It Dv IPV6_PORTRANGE_HIGH 266Use a high range (varies, see 267.Xr sysctl 8 ) . 268.It Dv IPV6_PORTRANGE_LOW 269Use a low, reserved range (600\-1023). 270.El 271.It Dv IPV6_PKTINFO Fa "int *" 272Get or set whether additional information about subsequent packets will 273be provided as ancillary data along with the payload in subsequent 274.Xr recvmsg 2 275calls. 276The information is stored in the following structure in the ancillary 277data returned: 278.Bd -literal 279struct in6_pktinfo { 280 struct in6_addr ipi6_addr; /* src/dst IPv6 address */ 281 unsigned int ipi6_ifindex; /* send/recv if index */ 282}; 283.Ed 284.It Dv IPV6_HOPLIMIT Fa "int *" 285Get or set whether the hop limit header field from subsequent packets 286will be provided as ancillary data along with the payload in subsequent 287.Xr recvmsg 2 288calls. 289The value is stored as an 290.Vt int 291in the ancillary data returned. 292.\" .It Dv IPV6_NEXTHOP Fa "int *" 293.\" Get or set whether the address of the next hop for subsequent 294.\" packets will be provided as ancillary data along with the payload in 295.\" subsequent 296.\" .Xr recvmsg 2 297.\" calls. 298.\" The option is stored as a 299.\" .Vt sockaddr 300.\" structure in the ancillary data returned. 301.\" .Pp 302.\" This option requires superuser privileges. 303.It Dv IPV6_HOPOPTS Fa "int *" 304Get or set whether the hop-by-hop options from subsequent packets will be 305provided as ancillary data along with the payload in subsequent 306.Xr recvmsg 2 307calls. 308The option is stored in the following structure in the ancillary data 309returned: 310.Bd -literal 311struct ip6_hbh { 312 uint8_t ip6h_nxt; /* next header */ 313 uint8_t ip6h_len; /* length in units of 8 octets */ 314/* followed by options */ 315} __packed; 316.Ed 317.Pp 318The 319.Fn inet6_option_space 320routine and family of routines may be used to manipulate this data. 321.Pp 322This option requires superuser privileges. 323.It Dv IPV6_DSTOPTS Fa "int *" 324Get or set whether the destination options from subsequent packets will 325be provided as ancillary data along with the payload in subsequent 326.Xr recvmsg 2 327calls. 328The option is stored in the following structure in the ancillary data 329returned: 330.Bd -literal 331struct ip6_dest { 332 uint8_t ip6d_nxt; /* next header */ 333 uint8_t ip6d_len; /* length in units of 8 octets */ 334/* followed by options */ 335} __packed; 336.Ed 337.Pp 338The 339.Fn inet6_option_space 340routine and family of routines may be used to manipulate this data. 341.Pp 342This option requires superuser privileges. 343.It Dv IPV6_RTHDR Fa "int *" 344Get or set whether the routing header from subsequent packets will be 345provided as ancillary data along with the payload in subsequent 346.Xr recvmsg 2 347calls. 348The header is stored in the following structure in the ancillary data 349returned: 350.Bd -literal 351struct ip6_rthdr { 352 uint8_t ip6r_nxt; /* next header */ 353 uint8_t ip6r_len; /* length in units of 8 octets */ 354 uint8_t ip6r_type; /* routing type */ 355 uint8_t ip6r_segleft; /* segments left */ 356/* followed by routing-type-specific data */ 357} __packed; 358.Ed 359.Pp 360The 361.Fn inet6_option_space 362routine and family of routines may be used to manipulate this data. 363.Pp 364This option requires superuser privileges. 365.It Dv IPV6_PKTOPTIONS Fa "struct cmsghdr *" 366Get or set all header options and extension headers at one time on the 367last packet sent or received on the socket. 368All options must fit within the size of an mbuf (see 369.Xr mbuf 9 ) . 370Options are specified as a series of 371.Vt cmsghdr 372structures followed by corresponding values. 373.Va cmsg_level 374is set to 375.Dv IPPROTO_IPV6 , 376.Va cmsg_type 377to one of the other values in this list, and trailing data to the option 378value. 379When setting options, if the length 380.Va optlen 381to 382.Xr setsockopt 2 383is zero, all header options will be reset to their default values. 384Otherwise, the length should specify the size the series of control 385messages consumes. 386.Pp 387Instead of using 388.Xr sendmsg 2 389to specify option values, the ancillary data used in these calls that 390correspond to the desired header options may be directly specified as 391the control message in the series of control messages provided as the 392argument to 393.Xr setsockopt 2 . 394.It Dv IPV6_CHECKSUM Fa "int *" 395Get or set the byte offset into a packet where the 16-bit checksum is 396located. 397When set, this byte offset is where incoming packets will be expected 398to have checksums of their data stored and where outgoing packets will 399have checksums of their data computed and stored by the kernel. 400A value of \-1 specifies that no checksums will be checked on incoming 401packets and that no checksums will be computed or stored on outgoing 402packets. 403The offset of the checksum for ICMPv6 sockets cannot be relocated or 404turned off. 405.It Dv IPV6_V6ONLY Fa "int *" 406Get or set whether only IPv6 connections can be made to this socket. 407For wildcard sockets, this can restrict connections to IPv6 only. 408.\"With 409.\".Ox 410.\"IPv6 sockets are always IPv6-only, so the socket option is read-only 411.\"(not modifiable). 412.It Dv IPV6_FAITH Fa "int *" 413Get or set the status of whether 414.Xr faith 4 415connections can be made to this socket. 416.It Dv IPV6_USE_MIN_MTU Fa "int *" 417Get or set whether the minimal IPv6 maximum transmission unit (MTU) size 418will be used to avoid fragmentation from occurring for subsequent 419outgoing datagrams. 420.It Dv IPV6_AUTH_LEVEL Fa "int *" 421Get or set the 422.Xr ipsec 4 423authentication level. 424.It Dv IPV6_ESP_TRANS_LEVEL Fa "int *" 425Get or set the ESP transport level. 426.It Dv IPV6_ESP_NETWORK_LEVEL Fa "int *" 427Get or set the ESP encapsulation level. 428.It Dv IPV6_IPCOMP_LEVEL Fa "int *" 429Get or set the 430.Xr ipcomp 4 431level. 432.It Dv IPV6_BINDANY 433If this option is enabled on a 434.Dv SOCK_STREAM , 435.Dv SOCK_DGRAM , 436or a 437.Dv SOCK_RAW 438socket, one can 439.Xr bind 2 440to any address, even one not bound to any available network interface in the 441system. 442This functionality (in conjunction with special firewall rules) can be used for 443implementing a transparent proxy. 444The 445.Dv KAUTH_REQ_NETWORK_BIND_ANYADDR 446privilege is needed to set this option. 447.El 448.Pp 449The 450.Dv IPV6_PKTINFO , 451.\" .Dv IPV6_NEXTHOP , 452.Dv IPV6_HOPLIMIT , 453.Dv IPV6_HOPOPTS , 454.Dv IPV6_DSTOPTS , 455and 456.Dv IPV6_RTHDR 457options will return ancillary data along with payload contents in subsequent 458.Xr recvmsg 2 459calls with 460.Va cmsg_level 461set to 462.Dv IPPROTO_IPV6 463and 464.Va cmsg_type 465set to respective option name value (e.g., 466.Dv IPV6_HOPTLIMIT ) . 467These options may also be used directly as ancillary 468.Va cmsg_type 469values in 470.Xr sendmsg 2 471to set options on the packet being transmitted by the call. 472The 473.Va cmsg_level 474value must be 475.Dv IPPROTO_IPV6 . 476For these options, the ancillary data object value format is the same 477as the value returned as explained for each when received with 478.Xr recvmsg 2 . 479.Pp 480Note that using 481.Xr sendmsg 2 482to specify options on particular packets works only on UDP and raw sockets. 483To manipulate header options for packets on TCP sockets, only the socket 484options may be used. 485.Pp 486In some cases, there are multiple APIs defined for manipulating an IPv6 487header field. 488A good example is the outgoing interface for multicast datagrams, which 489can be set by the 490.Dv IPV6_MULTICAST_IF 491socket option, through the 492.Dv IPV6_PKTINFO 493option, and through the 494.Va sin6_scope_id 495field of the socket address passed to the 496.Xr sendto 2 497system call. 498.Pp 499Resolving these conflicts is implementation dependent. 500This implementation determines the value in the following way: 501options specified by using ancillary data (i.e., 502.Xr sendmsg 2 ) 503are considered first, 504options specified by using 505.Dv IPV6_PKTOPTIONS 506to set 507.Dq sticky 508options are considered second, 509options specified by using the individual, basic, and direct socket 510options (e.g., 511.Dv IPV6_UNICAST_HOPS ) 512are considered third, 513and options specified in the socket address supplied to 514.Xr sendto 2 515are the last choice. 516.Ss Multicasting 517IPv6 multicasting is supported only on 518.Dv AF_INET6 519sockets of type 520.Dv SOCK_DGRAM 521and 522.Dv SOCK_RAW , 523and only on networks where the interface driver supports 524multicasting. 525Socket options (see above) that manipulate membership of 526multicast groups and other multicast options include 527.Dv IPV6_MULTICAST_IF , 528.Dv IPV6_MULTICAST_HOPS , 529.Dv IPV6_MULTICAST_LOOP , 530.Dv IPV6_LEAVE_GROUP , 531and 532.Dv IPV6_JOIN_GROUP . 533.Ss Raw Sockets 534Raw IPv6 sockets are connectionless and are normally used with the 535.Xr sendto 2 536and 537.Xr recvfrom 2 538calls, although the 539.Xr connect 2 540call may be used to fix the destination address for future outgoing 541packets so that 542.Xr send 2 543may instead be used and the 544.Xr bind 2 545call may be used to fix the source address for future outgoing 546packets instead of having the kernel choose a source address. 547.Pp 548By using 549.Xr connect 2 550or 551.Xr bind 2 , 552raw socket input is constrained to only packets with their 553source address matching the socket destination address if 554.Xr connect 2 555was used and to packets with their destination address 556matching the socket source address if 557.Xr bind 2 558was used. 559.Pp 560If the 561.Ar proto 562argument to 563.Xr socket 2 564is zero, the default protocol 565.Pq Dv IPPROTO_RAW 566is used for outgoing packets. 567For incoming packets, protocols recognized by kernel are 568.Sy not 569passed to the application socket (e.g., 570.Xr tcp 4 571and 572.Xr udp 4 ) 573except for some ICMPv6 messages. 574The ICMPv6 messages not passed to raw sockets include echo, timestamp, 575and address mask requests. 576If 577.Ar proto 578is non-zero, only packets with this protocol will be passed to the 579socket. 580.Pp 581IPv6 fragments are also not passed to application sockets until 582they have been reassembled. 583If reception of all packets is desired, link-level access (such as 584.Xr bpf 4 ) 585must be used instead. 586.Pp 587Outgoing packets automatically have an IPv6 header prepended to them 588(based on the destination address and the protocol number the socket 589was created with). 590Incoming packets are received by an application without the IPv6 header 591or any extension headers. 592.Pp 593Outgoing packets will be fragmented automatically by the kernel if they 594are too large. 595Incoming packets will be reassembled before being sent to the raw socket, 596so packet fragments or fragment headers will never be seen on a raw socket. 597.Sh EXAMPLES 598The following determines the hop limit on the next packet received: 599.Bd -literal 600struct iovec iov[2]; 601u_char buf[BUFSIZ]; 602struct cmsghdr *cm; 603struct msghdr m; 604int found, optval; 605u_char data[2048]; 606 607/* Create socket. */ 608 609(void)memset(&m, 0, sizeof(m)); 610(void)memset(&iov, 0, sizeof(iov)); 611 612iov[0].iov_base = data; /* buffer for packet payload */ 613iov[0].iov_len = sizeof(data); /* expected packet length */ 614 615m.msg_name = &from; /* sockaddr_in6 of peer */ 616m.msg_namelen = sizeof(from); 617m.msg_iov = iov; 618m.msg_iovlen = 1; 619m.msg_control = buf; /* buffer for control messages */ 620m.msg_controllen = sizeof(buf); 621 622/* 623 * Enable the hop limit value from received packets to be 624 * returned along with the payload. 625 */ 626optval = 1; 627if (setsockopt(s, IPPROTO_IPV6, IPV6_HOPLIMIT, &optval, 628 sizeof(optval)) == -1) 629 err(1, "setsockopt"); 630 631found = 0; 632while (!found) { 633 if (recvmsg(s, &m, 0) == -1) 634 err(1, "recvmsg"); 635 for (cm = CMSG_FIRSTHDR(&m); cm != NULL; 636 cm = CMSG_NXTHDR(&m, cm)) { 637 if (cm->cmsg_level == IPPROTO_IPV6 && 638 cm->cmsg_type == IPV6_HOPLIMIT && 639 cm->cmsg_len == CMSG_LEN(sizeof(int))) { 640 found = 1; 641 (void)printf("hop limit: %d\en", 642 *(int *)CMSG_DATA(cm)); 643 break; 644 } 645 } 646} 647.Ed 648.Sh DIAGNOSTICS 649A socket operation may fail with one of the following errors returned: 650.Bl -tag -width EADDRNOTAVAILxx 651.It Bq Er EISCONN 652when trying to establish a connection on a socket which 653already has one or when trying to send a datagram with the destination 654address specified and the socket is already connected. 655.It Bq Er ENOTCONN 656when trying to send a datagram, but 657no destination address is specified, and the socket hasn't been 658connected. 659.It Bq Er ENOBUFS 660when the system runs out of memory for 661an internal data structure. 662.It Bq Er EADDRNOTAVAIL 663when an attempt is made to create a 664socket with a network address for which no network interface 665exists. 666.It Bq Er EACCES 667when an attempt is made to create 668a raw IPv6 socket by a non-privileged process. 669.El 670.Pp 671The following errors specific to IPv6 may occur when setting or getting 672header options: 673.Bl -tag -width EADDRNOTAVAILxx 674.It Bq Er EINVAL 675An unknown socket option name was given. 676.It Bq Er EINVAL 677An ancillary data object was improperly formed. 678.El 679.Sh SEE ALSO 680.Xr getsockopt 2 , 681.Xr recv 2 , 682.Xr send 2 , 683.Xr setsockopt 2 , 684.Xr socket 2 , 685.Xr CMSG_DATA 3 , 686.\" .Xr inet6_option_space 3 , 687.\" .Xr inet6_rthdr_space 3 , 688.Xr if_nametoindex 3 , 689.Xr bpf 4 , 690.Xr icmp6 4 , 691.Xr inet6 4 , 692.Xr netintro 4 , 693.Xr tcp 4 , 694.Xr udp 4 695.Rs 696.%A W. Stevens 697.%A M. Thomas 698.%T Advanced Sockets API for IPv6 699.%R RFC 2292 700.%D February 1998 701.Re 702.Rs 703.%A S. Deering 704.%A R. Hinden 705.%T Internet Protocol, Version 6 (IPv6) Specification 706.%R RFC 2460 707.%D December 1998 708.Re 709.Rs 710.%A R. Gilligan 711.%A S. Thomson 712.%A J. Bound 713.%A W. Stevens 714.%T Basic Socket Interface Extensions for IPv6 715.%R RFC 2553 716.%D March 1999 717.Re 718.Rs 719.%A W. Stevens 720.%A B. Fenner 721.%A A. Rudoff 722.%T UNIX Network Programming, third edition 723.Re 724.Sh STANDARDS 725Most of the socket options are defined in RFC 2292 or RFC 2553. 726The 727.Dv IPV6_V6ONLY 728socket option is defined in RFC 3542. 729The 730.Dv IPV6_PORTRANGE 731socket option and the conflict resolution rule are not defined in the 732RFCs and should be considered implementation dependent. 733