xnu-2782.1.97.tar.gz

[apple/xnu.git] / bsd / netinet / tcp_ledbat.c

/*
 * Copyright (c) 2010-2014 Apple Inc. All rights reserved.
 *
 * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
 *
 * This file contains Original Code and/or Modifications of Original Code
 * as defined in and that are subject to the Apple Public Source License
 * Version 2.0 (the 'License'). You may not use this file except in
 * compliance with the License. The rights granted to you under the License
 * may not be used to create, or enable the creation or redistribution of,
 * unlawful or unlicensed copies of an Apple operating system, or to
 * circumvent, violate, or enable the circumvention or violation of, any
 * terms of an Apple operating system software license agreement.
 *
 * Please obtain a copy of the License at
 * http://www.opensource.apple.com/apsl/ and read it before using this file.
 *
 * The Original Code and all software distributed under the License are
 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
 * Please see the License for the specific language governing rights and
 * limitations under the License.
 *
 * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
 */
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/protosw.h>
#include <sys/mcache.h>
#include <sys/sysctl.h>

#include <net/route.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>

#if INET6
#include <netinet/ip6.h>
#endif
#include <netinet/ip_var.h>
#include <netinet/tcp.h>
#include <netinet/tcp_fsm.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcpip.h>
#include <netinet/tcp_cc.h>

#include <libkern/OSAtomic.h>

/* This file implements an alternate TCP congestion control algorithm
 * for background transport developed by LEDBAT working group at IETF and
 * described in draft: draft-ietf-ledbat-congestion-02
 */

int tcp_ledbat_init(struct tcpcb *tp);
int tcp_ledbat_cleanup(struct tcpcb *tp);
void tcp_ledbat_cwnd_init(struct tcpcb *tp);
void tcp_ledbat_congestion_avd(struct tcpcb *tp, struct tcphdr *th);
void tcp_ledbat_ack_rcvd(struct tcpcb *tp, struct tcphdr *th);
void tcp_ledbat_pre_fr(struct tcpcb *tp);
void tcp_ledbat_post_fr(struct tcpcb *tp, struct tcphdr *th);
void tcp_ledbat_after_idle(struct tcpcb *tp);
void tcp_ledbat_after_timeout(struct tcpcb *tp);
int tcp_ledbat_delay_ack(struct tcpcb *tp, struct tcphdr *th);
void tcp_ledbat_switch_cc(struct tcpcb *tp, uint16_t old_cc_index);

struct tcp_cc_algo tcp_cc_ledbat = {
        .name = "ledbat",
        .init = tcp_ledbat_init,
        .cleanup = tcp_ledbat_cleanup,
        .cwnd_init = tcp_ledbat_cwnd_init,
        .congestion_avd = tcp_ledbat_congestion_avd,
        .ack_rcvd = tcp_ledbat_ack_rcvd,
        .pre_fr = tcp_ledbat_pre_fr,
        .post_fr = tcp_ledbat_post_fr,
        .after_idle = tcp_ledbat_after_idle,
        .after_timeout = tcp_ledbat_after_timeout,
        .delay_ack = tcp_ledbat_delay_ack,
        .switch_to = tcp_ledbat_switch_cc
};

/* Target queuing delay in milliseconds. This includes the processing 
 * and scheduling delay on both of the end-hosts. A LEDBAT sender tries 
 * to keep queuing delay below this limit. When the queuing delay
 * goes above this limit, a LEDBAT sender will start reducing the 
 * congestion window.
 *
 * The LEDBAT draft says that target queue delay MUST be 100 ms for 
 * inter-operability.
 */
int target_qdelay = 100;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, bg_target_qdelay, CTLFLAG_RW | CTLFLAG_LOCKED, 
        &target_qdelay , 100, "Target queuing delay");

/* Allowed increase and tether are used to place an upper bound on
 * congestion window based on the amount of data that is outstanding.
 * This will limit the congestion window when the amount of data in 
 * flight is little because the application is writing to the socket
 * intermittently and is preventing the connection from becoming idle . 
 *
 * max_allowed_cwnd = allowed_increase + (tether * flight_size)
 * cwnd = min(cwnd, max_allowed_cwnd)
 *
 * 'Allowed_increase' parameter is set to 8. If the flight size is zero, then
 * we want the congestion window to be at least 8 packets to reduce the
 * delay induced by delayed ack. This helps when the receiver is acking 
 * more than 2 packets at a time (stretching acks for better performance).
 * 
 * 'Tether' is also set to 2. We do not want this to limit the growth of cwnd
 * during slow-start.
 */ 
int allowed_increase = 8;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, bg_allowed_increase, CTLFLAG_RW | CTLFLAG_LOCKED, 
        &allowed_increase, 1, "Additive constant used to calculate max allowed congestion window");

/* Left shift for cwnd to get tether value of 2 */
int tether_shift = 1;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, bg_tether_shift, CTLFLAG_RW | CTLFLAG_LOCKED, 
        &tether_shift, 1, "Tether shift for max allowed congestion window");

/* Start with an initial window of 2. This will help to get more accurate 
 * minimum RTT measurement in the beginning. It will help to probe
 * the path slowly and will not add to the existing delay if the path is
 * already congested. Using 2 packets will reduce the delay induced by delayed-ack.
 */
uint32_t bg_ss_fltsz = 2;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, bg_ss_fltsz, CTLFLAG_RW | CTLFLAG_LOCKED,
        &bg_ss_fltsz, 2, "Initial congestion window for background transport");

extern int rtt_samples_per_slot;

static void update_cwnd(struct tcpcb *tp, uint32_t incr) {
        uint32_t max_allowed_cwnd = 0, flight_size = 0;
        uint32_t qdelay, base_rtt;
        int32_t off_target;

        base_rtt = get_base_rtt(tp);

        /* If we do not have a good RTT measurement yet, increment
         * congestion window by the default value.  
         */
        if (base_rtt == 0 || tp->t_rttcur == 0) {
                tp->snd_cwnd += incr;
                goto check_max;
        }
                
        qdelay = tp->t_rttcur - base_rtt;
        off_target = (int32_t)(target_qdelay - qdelay);

        if (off_target >= 0) {
                /* Delay decreased or remained the same, we can increase 
                 * the congestion window according to RFC 3465.
                 *
                 * Move background slow-start threshold to current
                 * congestion window so that the next time (after some idle
                 * period), we can attempt to do slow-start till here if there 
                 * is no increase in rtt
                 */
                if (tp->bg_ssthresh < tp->snd_cwnd)
                        tp->bg_ssthresh = tp->snd_cwnd;
                tp->snd_cwnd += incr;   

        } else {
                /* In response to an increase in rtt, reduce the congestion 
                 * window by one-eighth. This will help to yield immediately 
                 * to a competing stream.
                 */
                uint32_t redwin;

                redwin = tp->snd_cwnd >> 3;  
                tp->snd_cwnd -= redwin;
                if (tp->snd_cwnd < bg_ss_fltsz * tp->t_maxseg)
                        tp->snd_cwnd = bg_ss_fltsz * tp->t_maxseg;

                /* Lower background slow-start threshold so that the connection 
                 * will go into congestion avoidance phase
                 */
                if (tp->bg_ssthresh > tp->snd_cwnd)
                        tp->bg_ssthresh = tp->snd_cwnd;
        }
check_max:
        /* Calculate the outstanding flight size and restrict the
         * congestion window to a factor of flight size.
         */
        flight_size = tp->snd_max - tp->snd_una;

        max_allowed_cwnd = (allowed_increase * tp->t_maxseg) 
                + (flight_size << tether_shift);
        tp->snd_cwnd = min(tp->snd_cwnd, max_allowed_cwnd);
        return;
}

int tcp_ledbat_init(struct tcpcb *tp) {
#pragma unused(tp)
        OSIncrementAtomic((volatile SInt32 *)&tcp_cc_ledbat.num_sockets);
        return 0;
}

int tcp_ledbat_cleanup(struct tcpcb *tp) {
#pragma unused(tp)
        OSDecrementAtomic((volatile SInt32 *)&tcp_cc_ledbat.num_sockets);
        return 0;
}

/* Initialize the congestion window for a connection 
 * 
 */

void
tcp_ledbat_cwnd_init(struct tcpcb *tp) {
        tp->snd_cwnd = tp->t_maxseg * bg_ss_fltsz;
        tp->bg_ssthresh = tp->snd_ssthresh;
}

/* Function to handle an in-sequence ack which is fast-path processing 
 * of an in sequence ack in tcp_input function (called as header prediction). 
 * This gets called only during congestion avoidance phase.
 */
void
tcp_ledbat_congestion_avd(struct tcpcb *tp, struct tcphdr *th) {
        int acked = 0;
        u_int32_t incr = 0;

        acked = BYTES_ACKED(th, tp);
        tp->t_bytes_acked += acked;
        if (tp->t_bytes_acked > tp->snd_cwnd) {
                tp->t_bytes_acked -= tp->snd_cwnd;
                incr = tp->t_maxseg;
        }

        if (tp->snd_cwnd < tp->snd_wnd && incr > 0) {
                update_cwnd(tp, incr);
        }
}
/* Function to process an ack.
 */
void
tcp_ledbat_ack_rcvd(struct tcpcb *tp, struct tcphdr *th) {
        /*
         * RFC 3465 - Appropriate Byte Counting.
         *
         * If the window is currently less than ssthresh,
         * open the window by the number of bytes ACKed by
         * the last ACK, however clamp the window increase
         * to an upper limit "L".
         *
         * In congestion avoidance phase, open the window by
         * one segment each time "bytes_acked" grows to be
         * greater than or equal to the congestion window.
         */

        register u_int cw = tp->snd_cwnd;
        register u_int incr = tp->t_maxseg;
        int acked = 0;

        acked = BYTES_ACKED(th, tp);
        tp->t_bytes_acked += acked;
        if (cw >= tp->bg_ssthresh) {
                /* congestion-avoidance */
                if (tp->t_bytes_acked < cw) {
                        /* No need to increase yet. */
                        incr = 0;
                }
        } else {
                /*
                 * If the user explicitly enables RFC3465
                 * use 2*SMSS for the "L" param.  Otherwise
                 * use the more conservative 1*SMSS.
                 *
                 * (See RFC 3465 2.3 Choosing the Limit)
                 */
                u_int abc_lim;

                abc_lim = (tcp_do_rfc3465_lim2 &&
                        tp->snd_nxt == tp->snd_max) ? incr * 2 : incr;

                incr = lmin(acked, abc_lim);
        }
        if (tp->t_bytes_acked >= cw)
                tp->t_bytes_acked -= cw;
        if (incr > 0) 
                update_cwnd(tp, incr);
}

void
tcp_ledbat_pre_fr(struct tcpcb *tp) {
        uint32_t win;

        win = min(tp->snd_wnd, tp->snd_cwnd) / 
                2 / tp->t_maxseg;
        if ( win < 2 )
                win = 2;
        tp->snd_ssthresh = win * tp->t_maxseg; 
        if (tp->bg_ssthresh > tp->snd_ssthresh)
                tp->bg_ssthresh = tp->snd_ssthresh;

        tcp_cc_resize_sndbuf(tp);
}

void
tcp_ledbat_post_fr(struct tcpcb *tp, struct tcphdr *th) {
        int32_t ss;

        ss = tp->snd_max - th->th_ack;

        /*
         * Complete ack.  Inflate the congestion window to
         * ssthresh and exit fast recovery.
         *
         * Window inflation should have left us with approx.
         * snd_ssthresh outstanding data.  But in case we
         * would be inclined to send a burst, better to do
         * it via the slow start mechanism.
         *
         * If the flight size is zero, then make congestion 
         * window to be worth at least 2 segments to avoid 
         * delayed acknowledgement (draft-ietf-tcpm-rfc3782-bis-05).
         */
        if (ss < (int32_t)tp->snd_ssthresh)
                tp->snd_cwnd = max(ss, tp->t_maxseg) + tp->t_maxseg;
        else
                tp->snd_cwnd = tp->snd_ssthresh;
        tp->t_bytes_acked = 0;
}

/*
 * Function to handle connections that have been idle for
 * some time. Slow start to get ack "clock" running again.
 * Clear base history after idle time.
 */
void
tcp_ledbat_after_idle(struct tcpcb *tp) {
        int32_t n = N_RTT_BASE, i = (N_RTT_BASE - 1);

        /* Decide how many base history entries have to be cleared 
         * based on how long the connection has been idle.
         */
        
        if (tp->t_rttcur > 0) {
                int32_t nrtt, idle_time;

                idle_time = tcp_now - tp->t_rcvtime;
                nrtt = idle_time / tp->t_rttcur; 
                n = nrtt / rtt_samples_per_slot;
                if (n > N_RTT_BASE)
                        n = N_RTT_BASE;
        }
        for (i = (N_RTT_BASE - 1); n > 0; --i, --n) {
                tp->rtt_hist[i] = 0;
        }
        for (n = (N_RTT_BASE - 1); i >= 0; --i, --n) {
                tp->rtt_hist[n] = tp->rtt_hist[i];
                tp->rtt_hist[i] = 0;
        }
        
        /* Reset the congestion window */
        tp->snd_cwnd = tp->t_maxseg * bg_ss_fltsz;

        /* If stretch ack was auto disabled, re-evaluate the situation */
        tcp_cc_after_idle_stretchack(tp);
}

/* Function to change the congestion window when the retransmit 
 * timer fires. The behavior is the same as that for best-effort
 * TCP, reduce congestion window to one segment and start probing
 * the link using "slow start". The slow start threshold is set
 * to half of the current window. Lower the background slow start
 * threshold also.
 */
void
tcp_ledbat_after_timeout(struct tcpcb *tp) {
        if (tp->t_state >=  TCPS_ESTABLISHED) {
                u_int win = min(tp->snd_wnd, tp->snd_cwnd) / 2 / tp->t_maxseg;
                if (win < 2)
                        win = 2;
                tp->snd_ssthresh = win * tp->t_maxseg;

                if (tp->bg_ssthresh > tp->snd_ssthresh)
                        tp->bg_ssthresh = tp->snd_ssthresh;

                tp->snd_cwnd = tp->t_maxseg;
                tcp_cc_resize_sndbuf(tp);
        }
}

/*
 * Indicate whether this ack should be delayed.
 * We can delay the ack if:
 *      - our last ack wasn't a 0-sized window.
 *      - the peer hasn't sent us a TH_PUSH data packet: if he did, take this 
 *      as a clue that we need to ACK without any delay. This helps higher 
 *      level protocols who won't send us more data even if the window is 
 *      open because their last "segment" hasn't been ACKed
 * Otherwise the receiver will ack every other full-sized segment or when the
 * delayed ack timer fires. This will help to generate better rtt estimates for 
 * the other end if it is a ledbat sender.
 * 
 */

int
tcp_ledbat_delay_ack(struct tcpcb *tp, struct tcphdr *th) {
        /* If any flag other than TH_ACK is set, set "end-of-write" bit */
        if (th->th_flags & ~TH_ACK)
                tp->t_flagsext |= TF_STREAMEOW;
        else
                tp->t_flagsext &= ~(TF_STREAMEOW);

        if ((tp->t_flags & TF_RXWIN0SENT) == 0 &&
                (th->th_flags & TH_PUSH) == 0 &&
                (tp->t_unacksegs == 1))
                return(1);
        return(0);
}

/* Change a connection to use ledbat. First, lower bg_ssthresh value
 * if it needs to be. 
 */
void
tcp_ledbat_switch_cc(struct tcpcb *tp, uint16_t old_cc_index) {
#pragma unused(old_cc_index)
        uint32_t cwnd;

        if (tp->bg_ssthresh == 0 || tp->bg_ssthresh > tp->snd_ssthresh)
                tp->bg_ssthresh = tp->snd_ssthresh;

        cwnd = min(tp->snd_wnd, tp->snd_cwnd);

        if (tp->snd_cwnd > tp->bg_ssthresh)
                cwnd = cwnd / tp->t_maxseg;
        else
                cwnd = cwnd / 2 / tp->t_maxseg;

        if (cwnd < bg_ss_fltsz)
                cwnd = bg_ss_fltsz;

        tp->snd_cwnd = cwnd * tp->t_maxseg;
        tp->t_bytes_acked = 0;

        OSIncrementAtomic((volatile SInt32 *)&tcp_cc_ledbat.num_sockets);
}