diff --git a/share/man/man4/tcp_bbr.4 b/share/man/man4/tcp_bbr.4 index 9944f6127ddd..fb1023dc81c6 100644 --- a/share/man/man4/tcp_bbr.4 +++ b/share/man/man4/tcp_bbr.4 @@ -1,167 +1,161 @@ .\" .\" Copyright (c) 2020, Gordon Bergling .\" .\" Redistribution and use in source and binary forms, with or without .\" modification, are permitted provided that the following conditions .\" are met: .\" 1. Redistributions of source code must retain the above copyright .\" notice, this list of conditions and the following disclaimer. .\" 2. Redistributions in binary form must reproduce the above copyright .\" notice, this list of conditions and the following disclaimer in the .\" documentation and/or other materials provided with the distribution. .\" .\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE .\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR .\" ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd November 7, 2022 +.Dd December 17, 2023 .Dt TCP_BBR 4 .Os .Sh NAME .Nm tcp_bbr .Nd TCP Bottleneck Bandwidth and Round-Trip Time Algorithm .Sh SYNOPSIS -To use this TCP stack you have to place the following line in your -kernel configuration file: -.Bd -ragged -offset indent -.Cd "options TCPHPTS" -.Ed -.Pp To load the driver as a module at boot time, place the following line in .Xr loader.conf 5 : .Bd -literal -offset indent tcp_bbr_load="YES" .Ed .Pp To enable the TCP stack you must place the following line in the .Xr sysctl.conf 5 : .Bd -literal -offset indent net.inet.tcp.functions_default=bbr .Ed .Sh DESCRIPTION Bottleneck bandwidth and round-trip time (BBR) is a congestion control algorithm which seeks high throughput with a small queue by probing BW and RTT. It is a round-up redesign of congestion control, which is not loss-based, delay-based, ECN-based or AIMD-based. .Pp The core design of BBR is about creating a model graph of the network path by estimating the maximum BW and minimum RTT on each ACK. .Sh MIB Variables The algorithm exposes the following scopes in the .Va net.inet.tcp.bbr branch of the .Xr sysctl 3 MIB: .Bl -tag -width ".Va exp_backoff_scale" .It Va cwnd Cwnd controls, for example "target cwnd rtt measurement" and "BBR initial window". .It Va measure Measurement controls. .It Va pacing Connection pacing controls. .It Va policer Policer controls, for example "false detection threshold" and "loss threshold". .It Va probertt Probe RTT controls. .It Va startup Startup controls. .It Va states State controls. .It Va timeout Time out controls. .El .Pp Besides the variables within the above scopes the following variables are also exposed in the .Va net.inet.tcp.bbr branch: .Bl -tag -width ".Va exp_backoff_scale" .It Va clrlost Clear lost counters. .It Va software_pacing Total number of software paced flows. .It Va hdwr_pacing Total number of hardware paced flows. .It Va enob_no_hdwr_pacing Total number of enobufs for non-hardware paced flows. .It Va enob_hdwr_pacing Total number of enobufs for hardware paced flows. .It Va rtt_tlp_thresh What divisor for TLP rtt/retran will be added (1=rtt, 2=1/2 rtt etc). .It Va reorder_fade Does reorder detection fade, if so how many ms (0 means never). .It Va reorder_thresh What factor for rack will be added when seeing reordering (shift right). .It Va bb_verbose Should BBR black box logging be verbose. .It Va sblklimit When do we start ignoring small sack blocks. .It Va resend_use_tso Can resends use TSO? .It Va data_after_close Do we hold off sending a RST until all pending data is ack'd. .It Va kill_paceout When we hit this many errors in a row, kill the session? .It Va error_paceout When we hit an error what is the min to pace out in usec's? .It Va cheat_rxt Do we burst 1ms between sends on retransmissions (like rack)? .It Va minrto Minimum RTO in ms. .El .Sh SEE ALSO .Xr cc_chd 4 , .Xr cc_cubic 4 , .Xr cc_hd 4 , .Xr cc_htcp 4 , .Xr cc_newreno 4 , .Xr cc_vegas 4 , .Xr h_ertt 4 , .Xr mod_cc 4 , .Xr tcp 4 , .Xr tcp_rack 4 , .Xr mod_cc 9 .Rs .%A "Neal Cardwell" .%A "Yuchung Cheng" .%A "Stephen Gunn" .%A "Soheil Hassas Yeganeh" .%A "Van Jacobson" .%T "BBR: Congestion-Based Congestion Control" .%J "ACM Queue, Vol. 14" .%D "September / October 2016" .Re .Rs .%A "Dominik Scholz" .%A "Benedikt Jaeger" .%A "Lukas Schwaighofer" .%A "Daniel Raumer" .%A "Fabien Geyer" .%A "Georg Carle" .%T "Towards a Deeper Understanding of TCP BBR Congestion Control" .%J "IFIP Networking 2018" .%D "May 2018" .%U "http://www.net.in.tum.de/fileadmin/bibtex/publications/papers/IFIP-Networking-2018-TCP-BBR.pdf" .Re .Sh HISTORY The .Nm congestion control module first appeared in .Fx 13.0 . .Sh AUTHORS .An -nosplit The .Nm congestion control module was written by .An Randall Stewart Aq Mt rrs@FreeBSD.org and sponsored by Netflix, Inc. This manual page was written by .An Gordon Bergling Aq Mt gbe@FreeBSD.org . diff --git a/share/man/man4/tcp_rack.4 b/share/man/man4/tcp_rack.4 index cfdedb9901b8..791b48baf6f6 100644 --- a/share/man/man4/tcp_rack.4 +++ b/share/man/man4/tcp_rack.4 @@ -1,159 +1,153 @@ .\" SPDX-License-Identifier: BSD-2-Clause .\" .\" Copyright (c) 2022, Gordon Bergling .\" .\" Redistribution and use in source and binary forms, with or without .\" modification, are permitted provided that the following conditions .\" are met: .\" 1. Redistributions of source code must retain the above copyright .\" notice, this list of conditions and the following disclaimer. .\" 2. Redistributions in binary form must reproduce the above copyright .\" notice, this list of conditions and the following disclaimer in the .\" documentation and/or other materials provided with the distribution. .\" .\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE .\" ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR .\" ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" -.Dd March 19, 2023 +.Dd December 17, 2023 .Dt TCP_RACK 4 .Os .Sh NAME .Nm tcp_rack .Nd TCP RACK-TLP Loss Detection Algorithm for TCP .Sh SYNOPSIS -To use this TCP stack, place the following line in the -kernel configuration file: -.Bd -ragged -offset indent -.Cd "options TCPHPTS" -.Ed -.Pp To load the TCP stack as a module at boot time, place the following line in .Xr loader.conf 5 : .Bd -literal -offset indent tcp_rack_load="YES" .Ed .Pp To enable the TCP stack, place the following line in the .Xr sysctl.conf 5 : .Bd -literal -offset indent net.inet.tcp.functions_default=rack .Ed .Sh DESCRIPTION RACK-TLP uses per-segment transmit timestamps and selective acknowledgments (SACKs) and has two parts. Recent Acknowledgment (RACK) starts fast recovery quickly using time-based inferences derived from acknowledgment (ACK) feedback, and Tail Loss Probe (TLP) leverages RACK and sends a probe packet to trigger ACK feedback to avoid retransmission timeout (RTO) events. .Pp Compared to the widely used duplicate acknowledgment (DupAck) threshold approach, RACK-TLP detects losses more efficiently when there are application-limited flights of data, lost retransmissions, or data packet reordering events. .Pp It is intended to be an alternative to the DupAck threshold approach. .Sh MIB Variables The algorithm exposes the following scopes in the .Va net.inet.tcp.rack branch of the .Xr sysctl 3 MIB: .Bl -tag -width indent .It Va net.inet.tcp.rack.misc Misc related controls .It Va net.inet.tcp.rack.features Feature controls .It Va net.inet.tcp.rack.measure Measure related controls .It Va net.inet.tcp.rack.timers Timer related controls .It Va net.inet.tcp.rack.tlp TLP and Rack related Controls .It Va net.inet.tcp.rack.timely Rack Timely RTT Controls .It Va net.inet.tcp.rack.hdwr_pacing Pacing related Controls .It Va net.inet.tcp.rack.pacing Pacing related Controls .It Va net.inet.tcp.rack.tp Rack tracepoint facility .It Va net.inet.tcp.rack.probertt ProbeRTT related Controls .It Va net.inet.tcp.rack.stats Rack Counters .It Va net.inet.tcp.rack.sack_attack Rack Sack Attack Counters and Controls .El .Pp Besides the variables within the above scopes the following variables are also exposed in the .Va net.inet.tcp.rack branch: .Bl -tag -width indent .It Va net.inet.tcp.rack.clear Clear counters .It Va net.inet.tcp.rack.opts RACK Option Stats .It Va net.inet.tcp.rack.outsize MSS send sizes .It Va net.inet.tcp.rack.req_measure_cnt If doing dynamic pacing, how many measurements must be in before we start pacing? .It Va net.inet.tcp.rack.use_pacing If set we use pacing, if clear we use only the original burst mitigation .It Va net.inet.tcp.rack.rate_sample_method What method should we use for rate sampling 0=high, 1=low .El .Sh SEE ALSO .Xr cc_chd 4 , .Xr cc_cubic 4 , .Xr cc_hd 4 , .Xr cc_htcp 4 , .Xr cc_newreno 4 , .Xr cc_vegas 4 , .Xr h_ertt 4 , .Xr mod_cc 4 , .Xr tcp 4 , .Xr tcp_bbr 4 , .Xr mod_cc 9 .Rs .%A "Neal Cardwell" .%A "Yuchung Cheng" .%A "Nandita Dukkipat" .%A "Priyaranjan Jha" .%T "The RACK-TLP Loss Detection Algorithm for TCP" .%O "RFC 8985" .%D "February 2021" .Re .Rs .%A "Ahmed M. Abdelmoniem" .%A "Brahim Bensaou" .%T "T-RACKs: A Faster Recovery Mechanism for TCP in Data Center Networks" .%O "https://arxiv.org/abs/2102.07477" .%D "February 2021" .Re .Sh HISTORY The .Nm congestion control module first appeared in .Fx 13.0 . .Sh AUTHORS .An -nosplit The .Nm congestion control module was written by .An Randall Stewart Aq Mt rrs@FreeBSD.org and sponsored by Netflix, Inc. This manual page was written by .An Gordon Bergling Aq Mt gbe@FreeBSD.org .