/* $Id: tcp_timer.c 106061 2024-09-16 14:03:52Z vboxsync $ */ /** @file * NAT - TCP timers. */ /* * Copyright (C) 2006-2024 Oracle and/or its affiliates. * * This file is part of VirtualBox base platform packages, as * available from https://www.virtualbox.org. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License * as published by the Free Software Foundation, in version 3 of the * License. * * This program is distributed in the hope that it will be useful, but * WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * General Public License for more details. * * You should have received a copy of the GNU General Public License * along with this program; if not, see . * * SPDX-License-Identifier: GPL-3.0-only */ /* * This code is based on: * * Copyright (c) 1982, 1986, 1988, 1990, 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. * 3. Neither the name of the University nor the names of its contributors * 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.c 8.1 (Berkeley) 6/10/93 * tcp_timer.c,v 1.2 1994/08/02 07:49:10 davidg Exp */ #include static struct tcpcb *tcp_timers(PNATState pData, register struct tcpcb *tp, int timer); /* * Fast timeout routine for processing delayed acks */ void tcp_fasttimo(PNATState pData) { register struct socket *so, *so_next; register struct tcpcb *tp; LogFlowFuncEnter(); so = tcb.so_next; if (so) QSOCKET_FOREACH (so, so_next, tcp) /* { */ if ( (tp = (struct tcpcb *)so->so_tcpcb) && (tp->t_flags & TF_DELACK)) { tp->t_flags &= ~TF_DELACK; tp->t_flags |= TF_ACKNOW; tcpstat.tcps_delack++; TCP_OUTPUT(pData, tp); } LOOP_LABEL(tcp, so, so_next); } } /* * Tcp protocol timeout routine called every 500 ms. * Updates the timers in all active tcb's and * causes finite state machine actions if timers expire. */ void tcp_slowtimo(PNATState pData) { register struct socket *ip, *ipnxt; register struct tcpcb *tp; register int i; LogFlowFuncEnter(); /* * Search through tcb's and update active timers. */ ip = tcb.so_next; if (ip == 0) return; QSOCKET_FOREACH(ip, ipnxt, tcp) /* { */ ipnxt = ip->so_next; tp = sototcpcb(ip); if (tp == 0) CONTINUE(tcp); for (i = 0; i < TCPT_NTIMERS; i++) { if (tp->t_timer[i] && --tp->t_timer[i] == 0) { tcp_timers(pData, tp, i); if (ipnxt->so_prev != ip) goto tpgone; } } tp->t_idle++; if (tp->t_rtt) tp->t_rtt++; tpgone: ; LOOP_LABEL(tcp, ip, ipnxt); } tcp_iss += TCP_ISSINCR / PR_SLOWHZ; /* increment iss */ #ifdef TCP_COMPAT_42 if ((int)tcp_iss < 0) tcp_iss = 0; /* XXX */ #endif tcp_now++; /* for timestamps */ } /* * Cancel all timers for TCP tp. */ void tcp_canceltimers(struct tcpcb *tp) { register int i; for (i = 0; i < TCPT_NTIMERS; i++) tp->t_timer[i] = 0; } const int tcp_backoff[TCP_MAXRXTSHIFT + 1] = { 1, 2, 4, 8, 16, 32, 64, 64, 64, 64, 64, 64, 64 }; /* * TCP timer processing. */ static struct tcpcb * tcp_timers(PNATState pData, register struct tcpcb *tp, int timer) { register int rexmt; int fUninitializedTemplate = 0; LogFlowFunc(("ENTER: tp:%R[tcpcb793], timer:%d\n", tp, timer)); fUninitializedTemplate = RT_BOOL(( tp->t_template.ti_src.s_addr == INADDR_ANY || tp->t_template.ti_dst.s_addr == INADDR_ANY)); if (fUninitializedTemplate) { tp = tcp_drop(pData, tp, 0); return tp; } switch (timer) { /* * 2 MSL timeout in shutdown went off. If we're closed but * still waiting for peer to close and connection has been idle * too long, or if 2MSL time is up from TIME_WAIT, delete connection * control block. Otherwise, check again in a bit. */ case TCPT_2MSL: if (tp->t_state != TCPS_TIME_WAIT && tp->t_idle <= tcp_maxidle) tp->t_timer[TCPT_2MSL] = tcp_keepintvl; else tp = tcp_close(pData, tp); break; /* * Retransmission timer went off. Message has not * been acked within retransmit interval. Back off * to a longer retransmit interval and retransmit one segment. */ case TCPT_REXMT: STAM_COUNTER_INC(&pData->StatTCP_retransmit); /* * XXX If a packet has timed out, then remove all the queued * packets for that session. */ if (++tp->t_rxtshift > TCP_MAXRXTSHIFT) { /* * This is a hack to suit our terminal server here at the uni of canberra * since they have trouble with zeroes... It usually lets them through * unharmed, but under some conditions, it'll eat the zeros. If we * keep retransmitting it, it'll keep eating the zeroes, so we keep * retransmitting, and eventually the connection dies... * (this only happens on incoming data) * * So, if we were gonna drop the connection from too many retransmits, * don't... instead halve the t_maxseg, which might break up the NULLs and * let them through * * *sigh* */ tp->t_maxseg >>= 1; if (tp->t_maxseg < 32) { /* * We tried our best, now the connection must die! */ tp->t_rxtshift = TCP_MAXRXTSHIFT; tcpstat.tcps_timeoutdrop++; tp = tcp_drop(pData, tp, tp->t_softerror); /* tp->t_softerror : ETIMEDOUT); */ /* XXX */ return (tp); /* XXX */ } /* * Set rxtshift to 6, which is still at the maximum * backoff time */ tp->t_rxtshift = 6; } tcpstat.tcps_rexmttimeo++; rexmt = TCP_REXMTVAL(tp) * tcp_backoff[tp->t_rxtshift]; TCPT_RANGESET(tp->t_rxtcur, rexmt, (short)tp->t_rttmin, TCPTV_REXMTMAX); /* XXX */ tp->t_timer[TCPT_REXMT] = tp->t_rxtcur; /* * If losing, let the lower level know and try for * a better route. Also, if we backed off this far, * our srtt estimate is probably bogus. Clobber it * so we'll take the next rtt measurement as our srtt; * move the current srtt into rttvar to keep the current * retransmit times until then. */ if (tp->t_rxtshift > TCP_MAXRXTSHIFT / 4) { /* in_losing(tp->t_inpcb); */ tp->t_rttvar += (tp->t_srtt >> TCP_RTT_SHIFT); tp->t_srtt = 0; } tp->snd_nxt = tp->snd_una; /* * If timing a segment in this window, stop the timer. */ tp->t_rtt = 0; /* * Close the congestion window down to one segment * (we'll open it by one segment for each ack we get). * Since we probably have a window's worth of unacked * data accumulated, this "slow start" keeps us from * dumping all that data as back-to-back packets (which * might overwhelm an intermediate gateway). * * There are two phases to the opening: Initially we * open by one mss on each ack. This makes the window * size increase exponentially with time. If the * window is larger than the path can handle, this * exponential growth results in dropped packet(s) * almost immediately. To get more time between * drops but still "push" the network to take advantage * of improving conditions, we switch from exponential * to linear window opening at some threshold size. * For a threshold, we use half the current window * size, truncated to a multiple of the mss. * * (the minimum cwnd that will give us exponential * growth is 2 mss. We don't allow the threshold * to go below this.) */ { u_int win = min(tp->snd_wnd, tp->snd_cwnd) / 2 / tp->t_maxseg; if (win < 2) win = 2; tp->snd_cwnd = tp->t_maxseg; tp->snd_ssthresh = win * tp->t_maxseg; tp->t_dupacks = 0; } (void) tcp_output(pData, tp); break; /* * Persistence timer into zero window. * Force a byte to be output, if possible. */ case TCPT_PERSIST: tcpstat.tcps_persisttimeo++; tcp_setpersist(tp); tp->t_force = 1; (void) tcp_output(pData, tp); tp->t_force = 0; break; /* * Keep-alive timer went off; send something * or drop connection if idle for too long. */ case TCPT_KEEP: tcpstat.tcps_keeptimeo++; if (tp->t_state < TCPS_ESTABLISHED) goto dropit; /* if (tp->t_socket->so_options & SO_KEEPALIVE && */ if ((so_options) && tp->t_state <= TCPS_CLOSE_WAIT) { if (tp->t_idle >= tcp_keepidle + tcp_maxidle) goto dropit; /* * Send a packet designed to force a response * if the peer is up and reachable: * either an ACK if the connection is still alive, * or an RST if the peer has closed the connection * due to timeout or reboot. * Using sequence number tp->snd_una-1 * causes the transmitted zero-length segment * to lie outside the receive window; * by the protocol spec, this requires the * correspondent TCP to respond. */ tcpstat.tcps_keepprobe++; #ifdef TCP_COMPAT_42 /* * The keepalive packet must have nonzero length * to get a 4.2 host to respond. */ tcp_respond(tp, &tp->t_template, (struct mbuf *)NULL, tp->rcv_nxt - 1, tp->snd_una - 1, 0); #else tcp_respond(pData, tp, &tp->t_template, (struct mbuf *)NULL, tp->rcv_nxt, tp->snd_una - 1, 0); #endif tp->t_timer[TCPT_KEEP] = tcp_keepintvl; } else tp->t_timer[TCPT_KEEP] = tcp_keepidle; break; dropit: tcpstat.tcps_keepdrops++; tp = tcp_drop(pData, tp, 0); /* ETIMEDOUT); */ break; } return tp; }