d/freebsd-nq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
c0029546f8
freebsd-nq/sys/netinet/tcp_timer.h

232 lines
9.2 KiB
C
Raw Normal View History

/* -> /*- for license, minor formatting changes
2005-01-07 01:45:51 +00:00
/*-
sys: further adoption of SPDX licensing ID tags. Mainly focus on files that use BSD 3-Clause license. The Software Package Data Exchange (SPDX) group provides a specification to make it easier for automated tools to detect and summarize well known opensource licenses. We are gradually adopting the specification, noting that the tags are considered only advisory and do not, in any way, superceed or replace the license texts. Special thanks to Wind River for providing access to "The Duke of Highlander" tool: an older (2014) run over FreeBSD tree was useful as a starting point.
2017-11-20 19:43:44 +00:00
* SPDX-License-Identifier: BSD-3-Clause
*
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
* Copyright (c) 1982, 1986, 1993
* The Regents of the University of California. All rights reserved.
*
* 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.
Renumber copyright clause 4 Renumber cluase 4 to 3, per what everybody else did when BSD granted them permission to remove clause 3. My insistance on keeping the same numbering for legal reasons is too pedantic, so give up on that point. Submitted by: Jan Schaumann <jschauma@stevens.edu> Pull Request: https://github.com/freebsd/freebsd/pull/96
2017-02-28 23:42:47 +00:00
* 3. Neither the name of the University nor the names of its contributors
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS 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 REGENTS 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.
*
* @(#)tcp_timer.h 8.1 (Berkeley) 6/10/93
$Id$ -> $FreeBSD$
1999-08-28 01:08:13 +00:00
* $FreeBSD$
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
*/
Made idempotent. Submitted by: Paul
1994-08-21 05:27:42 +00:00
#ifndef _NETINET_TCP_TIMER_H_
#define _NETINET_TCP_TIMER_H_
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* The TCPT_REXMT timer is used to force retransmissions.
* The TCP has the TCPT_REXMT timer set whenever segments
* have been sent for which ACKs are expected but not yet
* received. If an ACK is received which advances tp->snd_una,
* then the retransmit timer is cleared (if there are no more
* outstanding segments) or reset to the base value (if there
* are more ACKs expected). Whenever the retransmit timer goes off,
* we retransmit one unacknowledged segment, and do a backoff
* on the retransmit timer.
*
* The TCPT_PERSIST timer is used to keep window size information
* flowing even if the window goes shut. If all previous transmissions
* have been acknowledged (so that there are no retransmissions in progress),
* and the window is too small to bother sending anything, then we start
* the TCPT_PERSIST timer. When it expires, if the window is nonzero,
* we go to transmit state. Otherwise, at intervals send a single byte
* into the peer's window to force him to update our window information.
* We do this at most as often as TCPT_PERSMIN time intervals,
* but no more frequently than the current estimate of round-trip
* packet time. The TCPT_PERSIST timer is cleared whenever we receive
* a window update from the peer.
*
* The TCPT_KEEP timer is used to keep connections alive. If an
* connection is idle (no segments received) for TCPTV_KEEP_INIT amount of time,
* but not yet established, then we drop the connection. Once the connection
* is established, if the connection is idle for TCPTV_KEEP_IDLE time
* (and keepalives have been enabled on the socket), we begin to probe
* the connection. We force the peer to send us a segment by sending:
* <SEQ=SND.UNA-1><ACK=RCV.NXT><CTL=ACK>
* This segment is (deliberately) outside the window, and should elicit
* an ack segment in response from the peer. If, despite the TCPT_KEEP
* initiated segments we cannot elicit a response from a peer in TCPT_MAXIDLE
* amount of time probing, then we drop the connection.
*/
/*
* Time constants.
*/
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
#define TCPTV_MSL ( 30*hz) /* max seg lifetime (hah!) */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#define TCPTV_SRTTBASE 0 /* base roundtrip time;
if 0, no idea yet */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
#define TCPTV_RTOBASE ( 3*hz) /* assumed RTO if no info */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
Persist timers TCPTV_PERSMIN and TCPTV_PERSMAX are hardcoded with 5 seconds and 60 seconds, respectively. Turn them into sysctls that can be tuned live. The default values of 5 seconds and 60 seconds have been retained. Submitted by: Jason Wolfe (j at nitrology dot com) Reviewed by: gnn, rrs, hiren, bz MFC after: 1 week Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D5024
2016-01-26 16:33:38 +00:00
#define TCPTV_PERSMIN ( 5*hz) /* minimum persist interval */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
#define TCPTV_PERSMAX ( 60*hz) /* maximum persist interval */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
#define TCPTV_KEEP_INIT ( 75*hz) /* initial connect keepalive */
#define TCPTV_KEEP_IDLE (120*60*hz) /* dflt time before probing */
#define TCPTV_KEEPINTVL ( 75*hz) /* default probe interval */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#define TCPTV_KEEPCNT 8 /* max probes before drop */
Reap FIN_WAIT_2 connections marked SOCANTRCVMORE faster. This mitigate potential issues where the peer does not close, potentially leaving thousands of connections in FIN_WAIT_2. This is controlled by a new sysctl fast_finwait2_recycle, which is disabled by default. Reviewed by: gnn, silby.
2007-02-26 22:25:21 +00:00
#define TCPTV_FINWAIT2_TIMEOUT (60*hz) /* FIN_WAIT_2 timeout if no receiver */
I don't know how the minimum retransmit timeout managed to get set to one second but it badly breaks throughput on networks with minor packet loss. Complaints by: at least two people tracked down to this. MFC after: 3 days
2002-07-17 23:32:03 +00:00
/*
* Minimum retransmit timer is 3 ticks, for algorithmic stability.
Introduce two new sysctl's: net.inet.tcp.rexmit_min (default 3 ticks equiv) This sysctl is the retransmit timer RTO minimum, specified in milliseconds. This value is designed for algorithmic stability only. net.inet.tcp.rexmit_slop (default 200ms) This sysctl is the retransmit timer RTO slop which is added to every retransmit timeout and is designed to handle protocol stack overheads and delayed ack issues. Note that the *original* code applied a 1-second RTO minimum but never applied real slop to the RTO calculation, so any RTO calculation over one second would have no slop and thus not account for protocol stack overheads (TCP timestamps are not a measure of protocol turnaround!). Essentially, the original code made the RTO calculation almost completely irrelevant. Please note that the 200ms slop is debateable. This commit is not meant to be a line in the sand, and if the community winds up deciding that increasing it is the correct solution then it's easy to do. Note that larger values will destroy performance on lossy networks while smaller values may result in a greater number of unnecessary retransmits.
2002-07-18 19:06:12 +00:00
* TCPT_RANGESET() will add another TCPTV_CPU_VAR to deal with
* the expected worst-case processing variances by the kernels
* representing the end points. Such variances do not always show
* up in the srtt because the timestamp is often calculated at
* the interface rather then at the TCP layer. This value is
* typically 50ms. However, it is also possible that delayed
* acks (typically 100ms) could create issues so we set the slop
* to 200ms to try to cover it. Note that, properly speaking,
* delayed-acks should not create a major issue for interactive
White space cleanup for netinet before branch: - Trailing tab/space cleanup - Remove spurious spaces between or before tabs This change avoids touching files that Andre likely has in his working set for PFIL hooks changes for IPFW/DUMMYNET. Approved by: re (scottl) Submitted by: Xin LI <delphij@frontfree.net>
2004-08-16 18:32:07 +00:00
* environments which 'P'ush the last segment, at least as
Introduce two new sysctl's: net.inet.tcp.rexmit_min (default 3 ticks equiv) This sysctl is the retransmit timer RTO minimum, specified in milliseconds. This value is designed for algorithmic stability only. net.inet.tcp.rexmit_slop (default 200ms) This sysctl is the retransmit timer RTO slop which is added to every retransmit timeout and is designed to handle protocol stack overheads and delayed ack issues. Note that the *original* code applied a 1-second RTO minimum but never applied real slop to the RTO calculation, so any RTO calculation over one second would have no slop and thus not account for protocol stack overheads (TCP timestamps are not a measure of protocol turnaround!). Essentially, the original code made the RTO calculation almost completely irrelevant. Please note that the 200ms slop is debateable. This commit is not meant to be a line in the sand, and if the community winds up deciding that increasing it is the correct solution then it's easy to do. Note that larger values will destroy performance on lossy networks while smaller values may result in a greater number of unnecessary retransmits.
2002-07-18 19:06:12 +00:00
* long as implementations do the required 'at least one ack
* for every two packets' for the non-interactive streaming case.
* (maybe the RTO calculation should use 2*RTT instead of RTT
* to handle the ack-every-other-packet case).
*
* The prior minimum of 1*hz (1 second) badly breaks throughput on any
* networks faster then a modem that has minor (e.g. 1%) packet loss.
I don't know how the minimum retransmit timeout managed to get set to one second but it badly breaks throughput on networks with minor packet loss. Complaints by: at least two people tracked down to this. MFC after: 3 days
2002-07-17 23:32:03 +00:00
*/
Change TCPTV_MIN to be independent of HZ. While it was documented to be in ticks "for algorithm stability" when originally committed, it turns out that it has a significant impact in timing out connections. When we changed HZ from 100 to 1000, this had a big effect on reducing the time before dropping connections. To demonstrate, boot with kern.hz=100. ssh to a box on local ethernet and establish a reliable round-trip-time (ie: type a few commands). Then unplug the ethernet and press a key. Time how long it takes to drop the connection. The old behavior (with hz=100) caused the connection to typically drop between 90 and 110 seconds of getting no response. Now boot with kern.hz=1000 (default). The same test causes the ssh session to drop after just 9-10 seconds. This is a big deal on a wifi connection. With kern.hz=1000, change sysctl net.inet.tcp.rexmit_min from 3 to 30. Note how it behaves the same as when HZ was 100. Also, note that when booting with hz=100, net.inet.tcp.rexmit_min *used* to be 30. This commit changes TCPTV_MIN to be scaled with hz. rexmit_min should always be about 30. If you set hz to Really Slow(TM), there is a safety feature to prevent a value of 0 being used. This may be revised in the future, but for the time being, it restores the old, pre-hz=1000 behavior, which is significantly less annoying. As a workaround, to avoid rebooting or rebuilding a kernel, you can run "sysctl net.inet.tcp.rexmit_min=30" and add "net.inet.tcp.rexmit_min=30" to /etc/sysctl.conf. This is safe to run from 6.0 onwards. Approved by: re (rwatson) Reviewed by: andre, silby
2007-07-31 22:11:55 +00:00
#define TCPTV_MIN ( hz/33 ) /* minimum allowable value */
Introduce two new sysctl's: net.inet.tcp.rexmit_min (default 3 ticks equiv) This sysctl is the retransmit timer RTO minimum, specified in milliseconds. This value is designed for algorithmic stability only. net.inet.tcp.rexmit_slop (default 200ms) This sysctl is the retransmit timer RTO slop which is added to every retransmit timeout and is designed to handle protocol stack overheads and delayed ack issues. Note that the *original* code applied a 1-second RTO minimum but never applied real slop to the RTO calculation, so any RTO calculation over one second would have no slop and thus not account for protocol stack overheads (TCP timestamps are not a measure of protocol turnaround!). Essentially, the original code made the RTO calculation almost completely irrelevant. Please note that the 200ms slop is debateable. This commit is not meant to be a line in the sand, and if the community winds up deciding that increasing it is the correct solution then it's easy to do. Note that larger values will destroy performance on lossy networks while smaller values may result in a greater number of unnecessary retransmits.
2002-07-18 19:06:12 +00:00
#define TCPTV_CPU_VAR ( hz/5 ) /* cpu variance allowed (200ms) */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
#define TCPTV_REXMTMAX ( 64*hz) /* max allowable REXMT value */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
Merge in T/TCP TCP header file changes.
1995-02-08 20:18:48 +00:00
#define TCPTV_TWTRUNC 8 /* RTO factor to truncate TW */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#define TCP_LINGERTIME 120 /* linger at most 2 minutes */
#define TCP_MAXRXTSHIFT 12 /* maximum retransmits */
Define the delayed ACK timeout value directly as hz/10 instead of obfuscating it by going through PR_FASTHZ. No functional change. MFC after: 2 weeks
2012-10-29 12:17:02 +00:00
#define TCPTV_DELACK ( hz/10 ) /* 100ms timeout */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
Don't zero out srtt after excess retransmits If the TCP stack has retransmitted more than 1/4 of the total number of retransmits before a connection drop, it decides that its current RTT estimate is hopelessly out of date and decides to recalculate it from scratch starting with the next ACK. Unfortunately, it implements this by zeroing out the current RTT estimate. Drop this hack entirely, as it makes it significantly more difficult to debug connection issues. Instead check for excessive retransmits at the point where srtt is updated from an ACK being received. If we've exceeded 1/4 of the maximum retransmits, discard the previous srtt estimate and replace it with the latest rtt measurement. Differential Revision: https://reviews.freebsd.org/D9519 Reviewed by: gnn Sponsored by: Dell EMC Isilon
2017-02-11 17:05:08 +00:00
/*
* If we exceed this number of retransmits for a single segment, we'll consider
* the current srtt measurement no longer valid and will recalculate from
* scratch starting with the next ACK.
*/
#define TCP_RTT_INVALIDATE (TCP_MAXRXTSHIFT / 4)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#ifdef TCPTIMERS
make the strings for tcptimers, tanames and prurequests const to silence warnings.
2002-08-16 09:07:59 +00:00
static const char *tcptimers[] =
Add DELACK to list of timers. MFC after: 1 week
2012-11-27 19:07:28 +00:00
{ "REXMT", "PERSIST", "KEEP", "2MSL", "DELACK" };
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#endif
Make "struct tcp_timer" visible only to the kernel, and unbreak world.
2007-04-11 14:08:42 +00:00
/*
* Force a time value to be in a certain range.
*/
#define TCPT_RANGESET(tv, value, tvmin, tvmax) do { \
(tv) = (value) + tcp_rexmit_slop; \
if ((u_long)(tv) < (u_long)(tvmin)) \
(tv) = (tvmin); \
if ((u_long)(tv) > (u_long)(tvmax)) \
(tv) = (tvmax); \
} while(0)
#ifdef _KERNEL
Add the ability to see TCP timers via netstat -x. This can be a useful feature when you have a seemingly stuck socket and want to figure out why it has not been closed yet. No plans to MFC this, as it changes the netstat sysctl ABI. Reviewed by: andre, rwatson, Eric Van Gyzen
2009-09-16 05:33:15 +00:00
struct xtcp_timer;
Two changes: - Reintegrate the ANSI C function declaration change from tcp_timer.c rev 1.92 - Reorganize the tcpcb structure so that it has a single pointer to the "tcp_timer" structure which contains all of the tcp timer callouts. This change means that when the single tcp timer change is reintegrated, tcpcb will not change in size, and therefore the ABI between netstat and the kernel will not change. Neither of these changes should have any functional impact. Reviewed by: bmah, rrs Approved by: re (bmah)
2007-09-24 05:26:24 +00:00
struct tcp_timer {
struct callout tt_rexmt; /* retransmit timer */
struct callout tt_persist; /* retransmit persistence */
struct callout tt_keep; /* keepalive */
struct callout tt_2msl; /* 2*msl TIME_WAIT timer */
struct callout tt_delack; /* delayed ACK timer */
Fix an old and well-documented use-after-free race condition in TCP timers: - Add a reference from tcpcb to its inpcb - Defer tcpcb deletion until TCP timers have finished Differential Revision: https://reviews.freebsd.org/D2079 Submitted by: jch, Marc De La Gueronniere <mdelagueronniere@verisign.com> Reviewed by: imp, rrs, adrian, jhb, bz Approved by: jhb Sponsored by: Verisign, Inc.
2015-04-16 10:00:06 +00:00
uint32_t tt_flags; /* Timers flags */
This cleans up the timers code in TCP to start using the new async_drain functionality. This as been tested in NF as well as by Verisign. Still to do in here is to remove all the old flags. They are currently left being maintained but probably are no longer needed. Sponsored by: Netflix Inc. Differential Revision: http://reviews.freebsd.org/D5924
2016-04-28 13:27:12 +00:00
uint32_t tt_draincnt; /* Count being drained */
Two changes: - Reintegrate the ANSI C function declaration change from tcp_timer.c rev 1.92 - Reorganize the tcpcb structure so that it has a single pointer to the "tcp_timer" structure which contains all of the tcp timer callouts. This change means that when the single tcp timer change is reintegrated, tcpcb will not change in size, and therefore the ABI between netstat and the kernel will not change. Neither of these changes should have any functional impact. Reviewed by: bmah, rrs Approved by: re (bmah)
2007-09-24 05:26:24 +00:00
};
Fix an old and well-documented use-after-free race condition in TCP timers: - Add a reference from tcpcb to its inpcb - Defer tcpcb deletion until TCP timers have finished Differential Revision: https://reviews.freebsd.org/D2079 Submitted by: jch, Marc De La Gueronniere <mdelagueronniere@verisign.com> Reviewed by: imp, rrs, adrian, jhb, bz Approved by: jhb Sponsored by: Verisign, Inc.
2015-04-16 10:00:06 +00:00
/*
* Flags for the tt_flags field.
*/
#define TT_DELACK 0x0001
#define TT_REXMT 0x0002
#define TT_PERSIST 0x0004
#define TT_KEEP 0x0008
#define TT_2MSL 0x0010
#define TT_MASK (TT_DELACK|TT_REXMT|TT_PERSIST|TT_KEEP|TT_2MSL)
This commit brings in a new refactored TCP stack called Rack. Rack includes the following features: - A different SACK processing scheme (the old sack structures are not used). - RACK (Recent acknowledgment) where counting dup-acks is no longer done instead time is used to knwo when to retransmit. (see the I-D) - TLP (Tail Loss Probe) where we will probe for tail-losses to attempt to try not to take a retransmit time-out. (see the I-D) - Burst mitigation using TCPHTPS - PRR (partial rate reduction) see the RFC. Once built into your kernel, you can select this stack by either socket option with the name of the stack is "rack" or by setting the global sysctl so the default is rack. Note that any connection that does not support SACK will be kicked back to the "default" base FreeBSD stack (currently known as "default"). To build this into your kernel you will need to enable in your kernel: makeoptions WITH_EXTRA_TCP_STACKS=1 options TCPHPTS Sponsored by: Netflix Inc. Differential Revision: https://reviews.freebsd.org/D15525
2018-06-07 18:18:13 +00:00
/*
* Suspend flags - used when suspending a timer
* from ever running again.
*/
#define TT_DELACK_SUS 0x0100
#define TT_REXMT_SUS 0x0200
#define TT_PERSIST_SUS 0x0400
#define TT_KEEP_SUS 0x0800
#define TT_2MSL_SUS 0x1000
Put r284245 back in place: If at first this fix was seen as a temporary workaround for a callout(9) issue, it turns out it is instead the right way to use callout in mpsafe mode without using callout_drain(). r284245 commit message: Fix a callout race condition introduced in TCP timers callouts with r281599. In TCP timer context, it is not enough to check callout_stop() return value to decide if a callout is still running or not, previous callout_reset() return values have also to be checked. Differential Revision: https://reviews.freebsd.org/D2763
2015-08-30 13:44:39 +00:00
Fix an old and well-documented use-after-free race condition in TCP timers: - Add a reference from tcpcb to its inpcb - Defer tcpcb deletion until TCP timers have finished Differential Revision: https://reviews.freebsd.org/D2079 Submitted by: jch, Marc De La Gueronniere <mdelagueronniere@verisign.com> Reviewed by: imp, rrs, adrian, jhb, bz Approved by: jhb Sponsored by: Verisign, Inc.
2015-04-16 10:00:06 +00:00
#define TT_STOPPED 0x00010000
Change the TCP timer system from using the callout system five times directly to a merged model where only one callout, the next to fire, is registered. Instead of callout_reset(9) and callout_stop(9) the new function tcp_timer_activate() is used which then internally manages the callout. The single new callout is a mutex callout on inpcb simplifying the locking a bit. tcp_timer() is the called function which handles all race conditions in one place and then dispatches the individual timer functions. Reviewed by: rwatson (earlier version)
2007-04-11 09:45:16 +00:00
Add new socket options: TCP_KEEPINIT, TCP_KEEPIDLE, TCP_KEEPINTVL and TCP_KEEPCNT, that allow to control initial timeout, idle time, idle re-send interval and idle send count on a per-socket basis. Reviewed by: andre, bz, lstewart
2012-02-05 16:53:02 +00:00
#define TP_KEEPINIT(tp) ((tp)->t_keepinit ? (tp)->t_keepinit : tcp_keepinit)
#define TP_KEEPIDLE(tp) ((tp)->t_keepidle ? (tp)->t_keepidle : tcp_keepidle)
#define TP_KEEPINTVL(tp) ((tp)->t_keepintvl ? (tp)->t_keepintvl : tcp_keepintvl)
#define TP_KEEPCNT(tp) ((tp)->t_keepcnt ? (tp)->t_keepcnt : tcp_keepcnt)
#define TP_MAXIDLE(tp) (TP_KEEPCNT(tp) * TP_KEEPINTVL(tp))
Persist timers TCPTV_PERSMIN and TCPTV_PERSMAX are hardcoded with 5 seconds and 60 seconds, respectively. Turn them into sysctls that can be tuned live. The default values of 5 seconds and 60 seconds have been retained. Submitted by: Jason Wolfe (j at nitrology dot com) Reviewed by: gnn, rrs, hiren, bz MFC after: 1 week Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D5024
2016-01-26 16:33:38 +00:00
extern int tcp_persmin; /* minimum persist interval */
extern int tcp_persmax; /* maximum persist interval */
Make the misnamed tcp initial keepalive timer value (which is really the time, in seconds, that state for non-established TCP sessions stays about) a sysctl modifyable variable. [part 1 of two commits, I just realized I can't play with the indices as I was typing this commit message.]
1996-09-13 23:51:44 +00:00
extern int tcp_keepinit; /* time to establish connection */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
extern int tcp_keepidle; /* time before keepalive probes begin */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
extern int tcp_keepintvl; /* time between keepalive probes */
Add new socket options: TCP_KEEPINIT, TCP_KEEPIDLE, TCP_KEEPINTVL and TCP_KEEPCNT, that allow to control initial timeout, idle time, idle re-send interval and idle send count on a per-socket basis. Reviewed by: andre, bz, lstewart
2012-02-05 16:53:02 +00:00
extern int tcp_keepcnt; /* number of keepalives */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
extern int tcp_delacktime; /* time before sending a delayed ACK */
extern int tcp_maxpersistidle;
Introduce two new sysctl's: net.inet.tcp.rexmit_min (default 3 ticks equiv) This sysctl is the retransmit timer RTO minimum, specified in milliseconds. This value is designed for algorithmic stability only. net.inet.tcp.rexmit_slop (default 200ms) This sysctl is the retransmit timer RTO slop which is added to every retransmit timeout and is designed to handle protocol stack overheads and delayed ack issues. Note that the *original* code applied a 1-second RTO minimum but never applied real slop to the RTO calculation, so any RTO calculation over one second would have no slop and thus not account for protocol stack overheads (TCP timestamps are not a measure of protocol turnaround!). Essentially, the original code made the RTO calculation almost completely irrelevant. Please note that the 200ms slop is debateable. This commit is not meant to be a line in the sand, and if the community winds up deciding that increasing it is the correct solution then it's easy to do. Note that larger values will destroy performance on lossy networks while smaller values may result in a greater number of unnecessary retransmits.
2002-07-18 19:06:12 +00:00
extern int tcp_rexmit_min;
extern int tcp_rexmit_slop;
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
extern int tcp_msl;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
extern int tcp_ttl; /* time to live for TCP segs */
extern int tcp_backoff[];
For retransmits of SYN|ACK from the syncache use the slightly more aggressive special tcp_syn_backoff[] retransmit schedule instead of the normal tcp_backoff[] schedule for established connections. MFC after: 2 weeks
2012-10-28 19:02:07 +00:00
extern int tcp_syn_backoff[];
This commit brings in a new refactored TCP stack called Rack. Rack includes the following features: - A different SACK processing scheme (the old sack structures are not used). - RACK (Recent acknowledgment) where counting dup-acks is no longer done instead time is used to knwo when to retransmit. (see the I-D) - TLP (Tail Loss Probe) where we will probe for tail-losses to attempt to try not to take a retransmit time-out. (see the I-D) - Burst mitigation using TCPHTPS - PRR (partial rate reduction) see the RFC. Once built into your kernel, you can select this stack by either socket option with the name of the stack is "rack" or by setting the global sysctl so the default is rack. Note that any connection that does not support SACK will be kicked back to the "default" base FreeBSD stack (currently known as "default"). To build this into your kernel you will need to enable in your kernel: makeoptions WITH_EXTRA_TCP_STACKS=1 options TCPHPTS Sponsored by: Netflix Inc. Differential Revision: https://reviews.freebsd.org/D15525
2018-06-07 18:18:13 +00:00
extern int tcp_totbackoff;
extern int tcp_rexmit_drop_options;
Made idempotent. Submitted by: Paul
1994-08-21 05:27:42 +00:00
Export tcp_always_keepalive for use by the Chelsio TOM module. This used to work by accident with ld.bfd even though always_keepalive was marked as static. LLD honors static more correctly, so export this variable properly (including moving it into the tcp_* namespace). Reviewed by: bz, emaste MFC after: 2 weeks Sponsored by: Chelsio Communications Differential Revision: https://reviews.freebsd.org/D14129
2018-01-30 23:01:37 +00:00
extern int tcp_always_keepalive;
Reap FIN_WAIT_2 connections marked SOCANTRCVMORE faster. This mitigate potential issues where the peer does not close, potentially leaving thousands of connections in FIN_WAIT_2. This is controlled by a new sysctl fast_finwait2_recycle, which is disabled by default. Reviewed by: gnn, silby.
2007-02-26 22:25:21 +00:00
extern int tcp_finwait2_timeout;
extern int tcp_fast_finwait2_recycle;
Declare pmtud_blackhole global variables in tcp_timer.h, so that alternative TCP stacks can legally use them.
2017-10-06 20:33:40 +00:00
VNET_DECLARE(int, tcp_pmtud_blackhole_detect);
#define V_tcp_pmtud_blackhole_detect VNET(tcp_pmtud_blackhole_detect)
VNET_DECLARE(int, tcp_pmtud_blackhole_mss);
#define V_tcp_pmtud_blackhole_mss VNET(tcp_pmtud_blackhole_mss)
VNET_DECLARE(int, tcp_v6pmtud_blackhole_mss);
#define V_tcp_v6pmtud_blackhole_mss VNET(tcp_v6pmtud_blackhole_mss)
This cleans up the timer code in TCP and also makes it so we do not take the INFO lock *unless* we are really going to delete the TCB. Differential Revision: D7136
2016-08-16 12:40:56 +00:00
void tcp_inpinfo_lock_del(struct inpcb *inp, struct tcpcb *tp);
Remove a panic(); if the zone allocator can't provide more timewait structures, reuse the oldest one. Also move the expiry timer from a per-structure callout to the tcp slow timer. Sponsored by: DARPA, NAI Labs
2003-03-08 22:06:20 +00:00
void tcp_timer_init(void);
Back out tcp_timer.c:1.93 and associated changes that reimplemented the many TCP timers as a single timer, but retain the API changes necessary to reintroduce this change. This will back out the source of at least two reported problems: lock leaks in certain timer edge cases, and TCP timers continuing to fire after a connection has closed (a bug previously fixed and then reintroduced with the timer rewrite). In a follow-up commit, some minor restylings and comment changes performed after the TCP timer rewrite will be reapplied, and a further change to allow the TCP timer rewrite to be added back without disturbing the ABI. The new design is believed to be a good thing, but the outstanding issues are leading to significant stability/correctness problems that are holding up 7.0. This patch was generated by silby, but is being committed by proxy due to poor network connectivity for silby this week. Approved by: re (kensmith) Submitted by: silby Tested by: rwatson, kris Problems reported by: peter, kris, others
2007-09-07 09:19:22 +00:00
void tcp_timer_2msl(void *xtp);
This cleans up the timers code in TCP to start using the new async_drain functionality. This as been tested in NF as well as by Verisign. Still to do in here is to remove all the old flags. They are currently left being maintained but probably are no longer needed. Sponsored by: Netflix Inc. Differential Revision: http://reviews.freebsd.org/D5924
2016-04-28 13:27:12 +00:00
void tcp_timer_discard(void *);
Move TIME_WAIT related functions and timer handling from files other than repo copied tcp_subr.c into tcp_timewait.c#1.284: tcp_input.c#1.350 tcp_timewait() -> tcp_twcheck() tcp_timer.c#1.92 tcp_timer_2msl_reset() -> tcp_tw_2msl_reset() tcp_timer.c#1.92 tcp_timer_2msl_stop() -> tcp_tw_2msl_stop() tcp_timer.c#1.92 tcp_timer_2msl_tw() -> tcp_tw_2msl_scan() This is a mechanical move with appropriate renames and making them static if used only locally. The tcp_tw_2msl_scan() cleanup function is still run from the tcp_slowtimo() in tcp_timer.c.
2007-05-16 17:14:25 +00:00
struct tcptw *
Fix a race condition in TCP timewait between tcp_tw_2msl_reuse() and tcp_tw_2msl_scan(). This race condition drives unplanned timewait timeout cancellation. Also simplify implementation by holding inpcb reference and removing tcptw reference counting. Differential Revision: https://reviews.freebsd.org/D826 Submitted by: Marc De la Gueronniere <mdelagueronniere@verisign.com> Submitted by: jch Reviewed By: jhb (mentor), adrian, rwatson Sponsored by: Verisign, Inc. MFC after: 2 weeks X-MFC-With: r264321
2014-10-30 08:53:56 +00:00
tcp_tw_2msl_scan(int reuse); /* XXX temporary? */
Back out tcp_timer.c:1.93 and associated changes that reimplemented the many TCP timers as a single timer, but retain the API changes necessary to reintroduce this change. This will back out the source of at least two reported problems: lock leaks in certain timer edge cases, and TCP timers continuing to fire after a connection has closed (a bug previously fixed and then reintroduced with the timer rewrite). In a follow-up commit, some minor restylings and comment changes performed after the TCP timer rewrite will be reapplied, and a further change to allow the TCP timer rewrite to be added back without disturbing the ABI. The new design is believed to be a good thing, but the outstanding issues are leading to significant stability/correctness problems that are holding up 7.0. This patch was generated by silby, but is being committed by proxy due to poor network connectivity for silby this week. Approved by: re (kensmith) Submitted by: silby Tested by: rwatson, kris Problems reported by: peter, kris, others
2007-09-07 09:19:22 +00:00
void tcp_timer_keep(void *xtp);
void tcp_timer_persist(void *xtp);
void tcp_timer_rexmt(void *xtp);
void tcp_timer_delack(void *xtp);
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
Change #ifdef KERNEL to #ifdef _KERNEL in the public headers. "KERNEL" is an application space macro and the applications are supposed to be free to use it as they please (but cannot). This is consistant with the other BSD's who made this change quite some time ago. More commits to come.
1999-12-29 04:46:21 +00:00
#endif /* _KERNEL */
Restructure TCP timeout handling: - eliminate the fast/slow timeout lists for TCP and instead use a callout entry for each timer. - increase the TCP timer granularity to HZ - implement "bad retransmit" recovery, as presented in "On Estimating End-to-End Network Path Properties", by Allman and Paxson. Submitted by: jlemon, wollmann
1999-08-30 21:17:07 +00:00
#endif /* !_NETINET_TCP_TIMER_H_ */
Reference in New Issue Copy Permalink