1.\" Copyright (c) 1983, 1991, 1993 2.\" The Regents of the University of California. 3.\" Copyright (c) 2010-2011 The FreeBSD Foundation 4.\" All rights reserved. 5.\" 6.\" Portions of this documentation were written at the Centre for Advanced 7.\" Internet Architectures, Swinburne University of Technology, Melbourne, 8.\" Australia by David Hayes under sponsorship from the FreeBSD Foundation. 9.\" 10.\" Redistribution and use in source and binary forms, with or without 11.\" modification, are permitted provided that the following conditions 12.\" are met: 13.\" 1. Redistributions of source code must retain the above copyright 14.\" notice, this list of conditions and the following disclaimer. 15.\" 2. Redistributions in binary form must reproduce the above copyright 16.\" notice, this list of conditions and the following disclaimer in the 17.\" documentation and/or other materials provided with the distribution. 18.\" 3. Neither the name of the University nor the names of its contributors 19.\" may be used to endorse or promote products derived from this software 20.\" without specific prior written permission. 21.\" 22.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 23.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 24.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 25.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 26.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 27.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 28.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 29.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 30.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 31.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 32.\" SUCH DAMAGE. 33.\" 34.\" From: @(#)tcp.4 8.1 (Berkeley) 6/5/93 35.\" $FreeBSD$ 36.\" 37.Dd December 19, 2020 38.Dt TCP 4 39.Os 40.Sh NAME 41.Nm tcp 42.Nd Internet Transmission Control Protocol 43.Sh SYNOPSIS 44.In sys/types.h 45.In sys/socket.h 46.In netinet/in.h 47.In netinet/tcp.h 48.Ft int 49.Fn socket AF_INET SOCK_STREAM 0 50.Sh DESCRIPTION 51The 52.Tn TCP 53protocol provides reliable, flow-controlled, two-way 54transmission of data. 55It is a byte-stream protocol used to 56support the 57.Dv SOCK_STREAM 58abstraction. 59.Tn TCP 60uses the standard 61Internet address format and, in addition, provides a per-host 62collection of 63.Dq "port addresses" . 64Thus, each address is composed 65of an Internet address specifying the host and network, 66with a specific 67.Tn TCP 68port on the host identifying the peer entity. 69.Pp 70Sockets utilizing the 71.Tn TCP 72protocol are either 73.Dq active 74or 75.Dq passive . 76Active sockets initiate connections to passive 77sockets. 78By default, 79.Tn TCP 80sockets are created active; to create a 81passive socket, the 82.Xr listen 2 83system call must be used 84after binding the socket with the 85.Xr bind 2 86system call. 87Only passive sockets may use the 88.Xr accept 2 89call to accept incoming connections. 90Only active sockets may use the 91.Xr connect 2 92call to initiate connections. 93.Pp 94Passive sockets may 95.Dq underspecify 96their location to match 97incoming connection requests from multiple networks. 98This technique, termed 99.Dq "wildcard addressing" , 100allows a single 101server to provide service to clients on multiple networks. 102To create a socket which listens on all networks, the Internet 103address 104.Dv INADDR_ANY 105must be bound. 106The 107.Tn TCP 108port may still be specified 109at this time; if the port is not specified, the system will assign one. 110Once a connection has been established, the socket's address is 111fixed by the peer entity's location. 112The address assigned to the 113socket is the address associated with the network interface 114through which packets are being transmitted and received. 115Normally, this address corresponds to the peer entity's network. 116.Pp 117.Tn TCP 118supports a number of socket options which can be set with 119.Xr setsockopt 2 120and tested with 121.Xr getsockopt 2 : 122.Bl -tag -width ".Dv TCP_FUNCTION_BLK" 123.It Dv TCP_INFO 124Information about a socket's underlying TCP session may be retrieved 125by passing the read-only option 126.Dv TCP_INFO 127to 128.Xr getsockopt 2 . 129It accepts a single argument: a pointer to an instance of 130.Vt "struct tcp_info" . 131.Pp 132This API is subject to change; consult the source to determine 133which fields are currently filled out by this option. 134.Fx 135specific additions include 136send window size, 137receive window size, 138and 139bandwidth-controlled window space. 140.It Dv TCP_CCALGOOPT 141Set or query congestion control algorithm specific parameters. 142See 143.Xr mod_cc 4 144for details. 145.It Dv TCP_CONGESTION 146Select or query the congestion control algorithm that TCP will use for the 147connection. 148See 149.Xr mod_cc 4 150for details. 151.It Dv TCP_FUNCTION_BLK 152Select or query the set of functions that TCP will use for this connection. 153This allows a user to select an alternate TCP stack. 154The alternate TCP stack must already be loaded in the kernel. 155To list the available TCP stacks, see 156.Va functions_available 157in the 158.Sx MIB Variables 159section further down. 160To list the default TCP stack, see 161.Va functions_default 162in the 163.Sx MIB Variables 164section. 165.It Dv TCP_KEEPINIT 166This 167.Xr setsockopt 2 168option accepts a per-socket timeout argument of 169.Vt "u_int" 170in seconds, for new, non-established 171.Tn TCP 172connections. 173For the global default in milliseconds see 174.Va keepinit 175in the 176.Sx MIB Variables 177section further down. 178.It Dv TCP_KEEPIDLE 179This 180.Xr setsockopt 2 181option accepts an argument of 182.Vt "u_int" 183for the amount of time, in seconds, that the connection must be idle 184before keepalive probes (if enabled) are sent for the connection of this 185socket. 186If set on a listening socket, the value is inherited by the newly created 187socket upon 188.Xr accept 2 . 189For the global default in milliseconds see 190.Va keepidle 191in the 192.Sx MIB Variables 193section further down. 194.It Dv TCP_KEEPINTVL 195This 196.Xr setsockopt 2 197option accepts an argument of 198.Vt "u_int" 199to set the per-socket interval, in seconds, between keepalive probes sent 200to a peer. 201If set on a listening socket, the value is inherited by the newly created 202socket upon 203.Xr accept 2 . 204For the global default in milliseconds see 205.Va keepintvl 206in the 207.Sx MIB Variables 208section further down. 209.It Dv TCP_KEEPCNT 210This 211.Xr setsockopt 2 212option accepts an argument of 213.Vt "u_int" 214and allows a per-socket tuning of the number of probes sent, with no response, 215before the connection will be dropped. 216If set on a listening socket, the value is inherited by the newly created 217socket upon 218.Xr accept 2 . 219For the global default see the 220.Va keepcnt 221in the 222.Sx MIB Variables 223section further down. 224.It Dv TCP_NODELAY 225Under most circumstances, 226.Tn TCP 227sends data when it is presented; 228when outstanding data has not yet been acknowledged, it gathers 229small amounts of output to be sent in a single packet once 230an acknowledgement is received. 231For a small number of clients, such as window systems 232that send a stream of mouse events which receive no replies, 233this packetization may cause significant delays. 234The boolean option 235.Dv TCP_NODELAY 236defeats this algorithm. 237.It Dv TCP_MAXSEG 238By default, a sender- and 239.No receiver- Ns Tn TCP 240will negotiate among themselves to determine the maximum segment size 241to be used for each connection. 242The 243.Dv TCP_MAXSEG 244option allows the user to determine the result of this negotiation, 245and to reduce it if desired. 246.It Dv TCP_NOOPT 247.Tn TCP 248usually sends a number of options in each packet, corresponding to 249various 250.Tn TCP 251extensions which are provided in this implementation. 252The boolean option 253.Dv TCP_NOOPT 254is provided to disable 255.Tn TCP 256option use on a per-connection basis. 257.It Dv TCP_NOPUSH 258By convention, the 259.No sender- Ns Tn TCP 260will set the 261.Dq push 262bit, and begin transmission immediately (if permitted) at the end of 263every user call to 264.Xr write 2 265or 266.Xr writev 2 . 267When this option is set to a non-zero value, 268.Tn TCP 269will delay sending any data at all until either the socket is closed, 270or the internal send buffer is filled. 271.It Dv TCP_MD5SIG 272This option enables the use of MD5 digests (also known as TCP-MD5) 273on writes to the specified socket. 274Outgoing traffic is digested; 275digests on incoming traffic are verified. 276When this option is enabled on a socket, all inbound and outgoing 277TCP segments must be signed with MD5 digests. 278.Pp 279One common use for this in a 280.Fx 281router deployment is to enable 282based routers to interwork with Cisco equipment at peering points. 283Support for this feature conforms to RFC 2385. 284.Pp 285In order for this option to function correctly, it is necessary for the 286administrator to add a tcp-md5 key entry to the system's security 287associations database (SADB) using the 288.Xr setkey 8 289utility. 290This entry can only be specified on a per-host basis at this time. 291.Pp 292If an SADB entry cannot be found for the destination, 293the system does not send any outgoing segments and drops any inbound segments. 294.It Dv TCP_STATS 295Manage collection of connection level statistics using the 296.Xr stats 3 297framework. 298.Pp 299Each dropped segment is taken into account in the TCP protocol statistics. 300.It Dv TCP_TXTLS_ENABLE 301Enable in-kernel Transport Layer Security (TLS) for data written to this 302socket. 303See 304.Xr ktls 4 305for more details. 306.It Dv TCP_TXTLS_MODE 307The integer argument can be used to get or set the current TLS transmit mode 308of a socket. 309See 310.Xr ktls 4 311for more details. 312.It Dv TCP_RXTLS_ENABLE 313Enable in-kernel TLS for data read from this socket. 314See 315.Xr ktls 4 316for more details. 317.It Dv TCP_REUSPORT_LB_NUMA 318Changes NUMA affinity filtering for an established TCP listen 319socket. 320This option takes a single integer argument which specifies 321the NUMA domain to filter on for this listen socket. 322The argument can also have the follwing special values: 323.Bl -tag -width "Dv TCP_REUSPORT_LB_NUMA" 324.It Dv TCP_REUSPORT_LB_NUMA_NODOM 325Remove NUMA filtering for this listen socket. 326.It Dv TCP_REUSPORT_LB_NUMA_CURDOM 327Filter traffic associated with the domain where the calling thread is 328currently executing. 329This is typically used after a process or thread inherits a listen 330socket from its parent, and sets its CPU affinity to a particular core. 331.El 332.El 333.Pp 334The option level for the 335.Xr setsockopt 2 336call is the protocol number for 337.Tn TCP , 338available from 339.Xr getprotobyname 3 , 340or 341.Dv IPPROTO_TCP . 342All options are declared in 343.In netinet/tcp.h . 344.Pp 345Options at the 346.Tn IP 347transport level may be used with 348.Tn TCP ; 349see 350.Xr ip 4 . 351Incoming connection requests that are source-routed are noted, 352and the reverse source route is used in responding. 353.Pp 354The default congestion control algorithm for 355.Tn TCP 356is 357.Xr cc_newreno 4 . 358Other congestion control algorithms can be made available using the 359.Xr mod_cc 4 360framework. 361.Ss MIB Variables 362The 363.Tn TCP 364protocol implements a number of variables in the 365.Va net.inet.tcp 366branch of the 367.Xr sysctl 3 368MIB. 369.Bl -tag -width ".Va TCPCTL_DO_RFC1323" 370.It Dv TCPCTL_DO_RFC1323 371.Pq Va rfc1323 372Implement the window scaling and timestamp options of RFC 1323 373(default is true). 374.It Dv TCPCTL_MSSDFLT 375.Pq Va mssdflt 376The default value used for the maximum segment size 377.Pq Dq MSS 378when no advice to the contrary is received from MSS negotiation. 379.It Dv TCPCTL_SENDSPACE 380.Pq Va sendspace 381Maximum 382.Tn TCP 383send window. 384.It Dv TCPCTL_RECVSPACE 385.Pq Va recvspace 386Maximum 387.Tn TCP 388receive window. 389.It Va log_in_vain 390Log any connection attempts to ports where there is not a socket 391accepting connections. 392The value of 1 limits the logging to 393.Tn SYN 394(connection establishment) packets only. 395That of 2 results in any 396.Tn TCP 397packets to closed ports being logged. 398Any value unlisted above disables the logging 399(default is 0, i.e., the logging is disabled). 400.It Va msl 401The Maximum Segment Lifetime, in milliseconds, for a packet. 402.It Va keepinit 403Timeout, in milliseconds, for new, non-established 404.Tn TCP 405connections. 406The default is 75000 msec. 407.It Va keepidle 408Amount of time, in milliseconds, that the connection must be idle 409before keepalive probes (if enabled) are sent. 410The default is 7200000 msec (2 hours). 411.It Va keepintvl 412The interval, in milliseconds, between keepalive probes sent to remote 413machines, when no response is received on a 414.Va keepidle 415probe. 416The default is 75000 msec. 417.It Va keepcnt 418Number of probes sent, with no response, before a connection 419is dropped. 420The default is 8 packets. 421.It Va always_keepalive 422Assume that 423.Dv SO_KEEPALIVE 424is set on all 425.Tn TCP 426connections, the kernel will 427periodically send a packet to the remote host to verify the connection 428is still up. 429.It Va icmp_may_rst 430Certain 431.Tn ICMP 432unreachable messages may abort connections in 433.Tn SYN-SENT 434state. 435.It Va do_tcpdrain 436Flush packets in the 437.Tn TCP 438reassembly queue if the system is low on mbufs. 439.It Va blackhole 440If enabled, disable sending of RST when a connection is attempted 441to a port where there is not a socket accepting connections. 442See 443.Xr blackhole 4 . 444.It Va delayed_ack 445Delay ACK to try and piggyback it onto a data packet. 446.It Va delacktime 447Maximum amount of time, in milliseconds, before a delayed ACK is sent. 448.It Va path_mtu_discovery 449Enable Path MTU Discovery. 450.It Va tcbhashsize 451Size of the 452.Tn TCP 453control-block hash table 454(read-only). 455This may be tuned using the kernel option 456.Dv TCBHASHSIZE 457or by setting 458.Va net.inet.tcp.tcbhashsize 459in the 460.Xr loader 8 . 461.It Va pcbcount 462Number of active process control blocks 463(read-only). 464.It Va syncookies 465Determines whether or not 466.Tn SYN 467cookies should be generated for outbound 468.Tn SYN-ACK 469packets. 470.Tn SYN 471cookies are a great help during 472.Tn SYN 473flood attacks, and are enabled by default. 474(See 475.Xr syncookies 4 . ) 476.It Va isn_reseed_interval 477The interval (in seconds) specifying how often the secret data used in 478RFC 1948 initial sequence number calculations should be reseeded. 479By default, this variable is set to zero, indicating that 480no reseeding will occur. 481Reseeding should not be necessary, and will break 482.Dv TIME_WAIT 483recycling for a few minutes. 484.It Va reass.cursegments 485The current total number of segments present in all reassembly queues. 486.It Va reass.maxsegments 487The maximum limit on the total number of segments across all reassembly 488queues. 489The limit can be adjusted as a tunable. 490.It Va reass.maxqueuelen 491The maximum number of segments allowed in each reassembly queue. 492By default, the system chooses a limit based on each TCP connection's 493receive buffer size and maximum segment size (MSS). 494The actual limit applied to a session's reassembly queue will be the lower of 495the system-calculated automatic limit and the user-specified 496.Va reass.maxqueuelen 497limit. 498.It Va rexmit_initial , rexmit_min , rexmit_slop 499Adjust the retransmit timer calculation for 500.Tn TCP . 501The slop is 502typically added to the raw calculation to take into account 503occasional variances that the 504.Tn SRTT 505(smoothed round-trip time) 506is unable to accommodate, while the minimum specifies an 507absolute minimum. 508While a number of 509.Tn TCP 510RFCs suggest a 1 511second minimum, these RFCs tend to focus on streaming behavior, 512and fail to deal with the fact that a 1 second minimum has severe 513detrimental effects over lossy interactive connections, such 514as a 802.11b wireless link, and over very fast but lossy 515connections for those cases not covered by the fast retransmit 516code. 517For this reason, we use 200ms of slop and a near-0 518minimum, which gives us an effective minimum of 200ms (similar to 519.Tn Linux ) . 520The initial value is used before an RTT measurement has been performed. 521.It Va initcwnd_segments 522Enable the ability to specify initial congestion window in number of segments. 523The default value is 10 as suggested by RFC 6928. 524Changing the value on fly would not affect connections using congestion window 525from the hostcache. 526Caution: 527This regulates the burst of packets allowed to be sent in the first RTT. 528The value should be relative to the link capacity. 529Start with small values for lower-capacity links. 530Large bursts can cause buffer overruns and packet drops if routers have small 531buffers or the link is experiencing congestion. 532.It Va newcwd 533Enable the New Congestion Window Validation mechanism as described in RFC 7661. 534This gently reduces the congestion window during periods, where TCP is 535application limited and the network bandwidth is not utilized completely. 536That prevents self-inflicted packet losses once the application starts to 537transmit data at a higher speed. 538.It Va rfc6675_pipe 539Calculate the bytes in flight using the algorithm described in RFC 6675, and 540is also a prerequisite to enable Proportional Rate Reduction. 541.It Va rfc3042 542Enable the Limited Transmit algorithm as described in RFC 3042. 543It helps avoid timeouts on lossy links and also when the congestion window 544is small, as happens on short transfers. 545.It Va rfc3390 546Enable support for RFC 3390, which allows for a variable-sized 547starting congestion window on new connections, depending on the 548maximum segment size. 549This helps throughput in general, but 550particularly affects short transfers and high-bandwidth large 551propagation-delay connections. 552.It Va sack.enable 553Enable support for RFC 2018, TCP Selective Acknowledgment option, 554which allows the receiver to inform the sender about all successfully 555arrived segments, allowing the sender to retransmit the missing segments 556only. 557.It Va sack.maxholes 558Maximum number of SACK holes per connection. 559Defaults to 128. 560.It Va sack.globalmaxholes 561Maximum number of SACK holes per system, across all connections. 562Defaults to 65536. 563.It Va maxtcptw 564When a TCP connection enters the 565.Dv TIME_WAIT 566state, its associated socket structure is freed, since it is of 567negligible size and use, and a new structure is allocated to contain a 568minimal amount of information necessary for sustaining a connection in 569this state, called the compressed TCP TIME_WAIT state. 570Since this structure is smaller than a socket structure, it can save 571a significant amount of system memory. 572The 573.Va net.inet.tcp.maxtcptw 574MIB variable controls the maximum number of these structures allocated. 575By default, it is initialized to 576.Va kern.ipc.maxsockets 577/ 5. 578.It Va nolocaltimewait 579Suppress creating of compressed TCP TIME_WAIT states for connections in 580which both endpoints are local. 581.It Va fast_finwait2_recycle 582Recycle 583.Tn TCP 584.Dv FIN_WAIT_2 585connections faster when the socket is marked as 586.Dv SBS_CANTRCVMORE 587(no user process has the socket open, data received on 588the socket cannot be read). 589The timeout used here is 590.Va finwait2_timeout . 591.It Va finwait2_timeout 592Timeout to use for fast recycling of 593.Tn TCP 594.Dv FIN_WAIT_2 595connections. 596Defaults to 60 seconds. 597.It Va ecn.enable 598Enable support for TCP Explicit Congestion Notification (ECN). 599ECN allows a TCP sender to reduce the transmission rate in order to 600avoid packet drops. 601Settings: 602.Bl -tag -compact 603.It 0 604Disable ECN. 605.It 1 606Allow incoming connections to request ECN. 607Outgoing connections will request ECN. 608.It 2 609Allow incoming connections to request ECN. 610Outgoing connections will not request ECN. 611.El 612.It Va ecn.maxretries 613Number of retries (SYN or SYN/ACK retransmits) before disabling ECN on a 614specific connection. 615This is needed to help with connection establishment 616when a broken firewall is in the network path. 617.It Va pmtud_blackhole_detection 618Enable automatic path MTU blackhole detection. 619In case of retransmits of MSS sized segments, 620the OS will lower the MSS to check if it's an MTU problem. 621If the current MSS is greater than the configured value to try 622.Po Va net.inet.tcp.pmtud_blackhole_mss 623and 624.Va net.inet.tcp.v6pmtud_blackhole_mss 625.Pc , 626it will be set to this value, otherwise, 627the MSS will be set to the default values 628.Po Va net.inet.tcp.mssdflt 629and 630.Va net.inet.tcp.v6mssdflt 631.Pc . 632Settings: 633.Bl -tag -compact 634.It 0 635Disable path MTU blackhole detection. 636.It 1 637Enable path MTU blackhole detection for IPv4 and IPv6. 638.It 2 639Enable path MTU blackhole detection only for IPv4. 640.It 3 641Enable path MTU blackhole detection only for IPv6. 642.El 643.It Va pmtud_blackhole_mss 644MSS to try for IPv4 if PMTU blackhole detection is turned on. 645.It Va v6pmtud_blackhole_mss 646MSS to try for IPv6 if PMTU blackhole detection is turned on. 647.It Va functions_available 648List of available TCP function blocks (TCP stacks). 649.It Va functions_default 650The default TCP function block (TCP stack). 651.It Va functions_inherit_listen_socket_stack 652Determines whether to inherit listen socket's tcp stack or use the current 653system default tcp stack, as defined by 654.Va functions_default . 655Default is true. 656.It Va insecure_rst 657Use criteria defined in RFC793 instead of RFC5961 for accepting RST segments. 658Default is false. 659.It Va insecure_syn 660Use criteria defined in RFC793 instead of RFC5961 for accepting SYN segments. 661Default is false. 662.It Va ts_offset_per_conn 663When initializing the TCP timestamps, use a per connection offset instead of a 664per host pair offset. 665Default is to use per connection offsets as recommended in RFC 7323. 666.It Va perconn_stats_enable 667Controls the default collection of statistics for all connections using the 668.Xr stats 3 669framework. 6700 disables, 1 enables, 2 enables random sampling across log id connection 671groups with all connections in a group receiving the same setting. 672.It Va perconn_stats_sample_rates 673A CSV list of template_spec=percent key-value pairs which controls the per 674template sampling rates when 675.Xr stats 3 676sampling is enabled. 677.El 678.Sh ERRORS 679A socket operation may fail with one of the following errors returned: 680.Bl -tag -width Er 681.It Bq Er EISCONN 682when trying to establish a connection on a socket which 683already has one; 684.It Bo Er ENOBUFS Bc or Bo Er ENOMEM Bc 685when the system runs out of memory for 686an internal data structure; 687.It Bq Er ETIMEDOUT 688when a connection was dropped 689due to excessive retransmissions; 690.It Bq Er ECONNRESET 691when the remote peer 692forces the connection to be closed; 693.It Bq Er ECONNREFUSED 694when the remote 695peer actively refuses connection establishment (usually because 696no process is listening to the port); 697.It Bq Er EADDRINUSE 698when an attempt 699is made to create a socket with a port which has already been 700allocated; 701.It Bq Er EADDRNOTAVAIL 702when an attempt is made to create a 703socket with a network address for which no network interface 704exists; 705.It Bq Er EAFNOSUPPORT 706when an attempt is made to bind or connect a socket to a multicast 707address. 708.It Bq Er EINVAL 709when trying to change TCP function blocks at an invalid point in the session; 710.It Bq Er ENOENT 711when trying to use a TCP function block that is not available; 712.El 713.Sh SEE ALSO 714.Xr getsockopt 2 , 715.Xr socket 2 , 716.Xr stats 3 , 717.Xr sysctl 3 , 718.Xr blackhole 4 , 719.Xr inet 4 , 720.Xr intro 4 , 721.Xr ip 4 , 722.Xr ktls 4 , 723.Xr mod_cc 4 , 724.Xr siftr 4 , 725.Xr syncache 4 , 726.Xr tcp_bbr 4 , 727.Xr setkey 8 , 728.Xr tcp_functions 9 729.Rs 730.%A "V. Jacobson" 731.%A "R. Braden" 732.%A "D. Borman" 733.%T "TCP Extensions for High Performance" 734.%O "RFC 1323" 735.Re 736.Rs 737.%A "A. Heffernan" 738.%T "Protection of BGP Sessions via the TCP MD5 Signature Option" 739.%O "RFC 2385" 740.Re 741.Rs 742.%A "K. Ramakrishnan" 743.%A "S. Floyd" 744.%A "D. Black" 745.%T "The Addition of Explicit Congestion Notification (ECN) to IP" 746.%O "RFC 3168" 747.Re 748.Sh HISTORY 749The 750.Tn TCP 751protocol appeared in 752.Bx 4.2 . 753The RFC 1323 extensions for window scaling and timestamps were added 754in 755.Bx 4.4 . 756The 757.Dv TCP_INFO 758option was introduced in 759.Tn Linux 2.6 760and is 761.Em subject to change . 762