| tcp_timer.h (128019) | tcp_timer.h (133874) |
|---|---|
| 1/* 2 * Copyright (c) 1982, 1986, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright --- 13 unchanged lines hidden (view full) --- 22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27 * SUCH DAMAGE. 28 * 29 * @(#)tcp_timer.h 8.1 (Berkeley) 6/10/93 | 1/* 2 * Copyright (c) 1982, 1986, 1993 3 * The Regents of the University of California. All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright --- 13 unchanged lines hidden (view full) --- 22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27 * SUCH DAMAGE. 28 * 29 * @(#)tcp_timer.h 8.1 (Berkeley) 6/10/93 |
| 30 * $FreeBSD: head/sys/netinet/tcp_timer.h 128019 2004-04-07 20:46:16Z imp $ | 30 * $FreeBSD: head/sys/netinet/tcp_timer.h 133874 2004-08-16 18:32:07Z rwatson $ |
| 31 */ 32 33#ifndef _NETINET_TCP_TIMER_H_ 34#define _NETINET_TCP_TIMER_H_ 35 36/* 37 * The TCPT_REXMT timer is used to force retransmissions. 38 * The TCP has the TCPT_REXMT timer set whenever segments --- 53 unchanged lines hidden (view full) --- 92 * the expected worst-case processing variances by the kernels 93 * representing the end points. Such variances do not always show 94 * up in the srtt because the timestamp is often calculated at 95 * the interface rather then at the TCP layer. This value is 96 * typically 50ms. However, it is also possible that delayed 97 * acks (typically 100ms) could create issues so we set the slop 98 * to 200ms to try to cover it. Note that, properly speaking, 99 * delayed-acks should not create a major issue for interactive | 31 */ 32 33#ifndef _NETINET_TCP_TIMER_H_ 34#define _NETINET_TCP_TIMER_H_ 35 36/* 37 * The TCPT_REXMT timer is used to force retransmissions. 38 * The TCP has the TCPT_REXMT timer set whenever segments --- 53 unchanged lines hidden (view full) --- 92 * the expected worst-case processing variances by the kernels 93 * representing the end points. Such variances do not always show 94 * up in the srtt because the timestamp is often calculated at 95 * the interface rather then at the TCP layer. This value is 96 * typically 50ms. However, it is also possible that delayed 97 * acks (typically 100ms) could create issues so we set the slop 98 * to 200ms to try to cover it. Note that, properly speaking, 99 * delayed-acks should not create a major issue for interactive |
| 100 * environments which 'P'ush the last segment, at least as | 100 * environments which 'P'ush the last segment, at least as |
| 101 * long as implementations do the required 'at least one ack 102 * for every two packets' for the non-interactive streaming case. 103 * (maybe the RTO calculation should use 2*RTT instead of RTT 104 * to handle the ack-every-other-packet case). 105 * 106 * The prior minimum of 1*hz (1 second) badly breaks throughput on any 107 * networks faster then a modem that has minor (e.g. 1%) packet loss. 108 */ --- 57 unchanged lines hidden --- | 101 * long as implementations do the required 'at least one ack 102 * for every two packets' for the non-interactive streaming case. 103 * (maybe the RTO calculation should use 2*RTT instead of RTT 104 * to handle the ack-every-other-packet case). 105 * 106 * The prior minimum of 1*hz (1 second) badly breaks throughput on any 107 * networks faster then a modem that has minor (e.g. 1%) packet loss. 108 */ --- 57 unchanged lines hidden --- |