diff -pruN linux-2.6.19.1-orig/net/ipv4/Kconfig linux-2.6.19.1/net/ipv4/Kconfig --- linux-2.6.19.1-orig/net/ipv4/Kconfig 2006-12-11 14:32:53.000000000 -0500 +++ linux-2.6.19.1/net/ipv4/Kconfig 2007-04-03 15:49:20.903349384 -0400 @@ -538,6 +538,12 @@ config TCP_CONG_HYBLA involved, especially when sharing a common bottleneck with normal terrestrial connections. +config TCP_CONG_AFRICA + tristate "TCP Africa" + depends on EXPERIMENTAL + default n + ---help--- + config TCP_CONG_VEGAS tristate "TCP Vegas" depends on EXPERIMENTAL diff -pruN linux-2.6.19.1-orig/net/ipv4/Makefile linux-2.6.19.1/net/ipv4/Makefile --- linux-2.6.19.1-orig/net/ipv4/Makefile 2006-12-11 14:32:53.000000000 -0500 +++ linux-2.6.19.1/net/ipv4/Makefile 2007-04-03 15:49:39.018595448 -0400 @@ -44,6 +44,7 @@ obj-$(CONFIG_TCP_CONG_WESTWOOD) += tcp_w obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o +obj-$(CONFIG_TCP_CONG_AFRICA) += tcp_africa.o obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o diff -pruN linux-2.6.19.1-orig/net/ipv4/tcp_africa.c linux-2.6.19.1/net/ipv4/tcp_africa.c --- linux-2.6.19.1-orig/net/ipv4/tcp_africa.c 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.19.1/net/ipv4/tcp_africa.c 2007-04-03 15:47:16.555253168 -0400 @@ -0,0 +1,297 @@ +/* -*- indent-tabs-mode: t; tab-width: 8; c-basic-offset: 8; -*- */ +/* + * TCP Africa congestion control + * + * This is based on the congestion detection/avoidance scheme described in + * King R., Baraniuk, R., and Riedi, R. + * "TCP-Africa: An Adaptive and Fair Rapid Increase Rule for Scalable TCP" + * Presented at INFOCOM 2005, and published in: + * INFOCOM 2005. 24th Annual Joint Conference of the IEEE Computer and + * Communications Societies. Proceedings IEEE + * Volume 3, 13-17 March 2005. pp 1838-1848. + * + * The paper is available from: + * http://www.spin.rice.edu/PDF/Infocom05web.pdf + * + * TCP-Africa is a delay-sensitive congestion avoidance algorithm for TCP that + * allows for scalable, aggressive behavior in large underutilized links, yet + * falls back to the more conservative TCP-Reno algorithm once links become + * well utilized and congestion is imminent. + * + */ + +#include +#include + +#include + +/* From AIMD tables from RFC 3649 appendix B, + * with fixed-point MD scaled <<8. + */ +static const struct hstcp_aimd_val { + unsigned int cwnd; + unsigned int md; +} hstcp_aimd_vals[] = { + { 38, 128, /* 0.50 */ }, + { 118, 112, /* 0.44 */ }, + { 221, 104, /* 0.41 */ }, + { 347, 98, /* 0.38 */ }, + { 495, 93, /* 0.37 */ }, + { 663, 89, /* 0.35 */ }, + { 851, 86, /* 0.34 */ }, + { 1058, 83, /* 0.33 */ }, + { 1284, 81, /* 0.32 */ }, + { 1529, 78, /* 0.31 */ }, + { 1793, 76, /* 0.30 */ }, + { 2076, 74, /* 0.29 */ }, + { 2378, 72, /* 0.28 */ }, + { 2699, 71, /* 0.28 */ }, + { 3039, 69, /* 0.27 */ }, + { 3399, 68, /* 0.27 */ }, + { 3778, 66, /* 0.26 */ }, + { 4177, 65, /* 0.26 */ }, + { 4596, 64, /* 0.25 */ }, + { 5036, 62, /* 0.25 */ }, + { 5497, 61, /* 0.24 */ }, + { 5979, 60, /* 0.24 */ }, + { 6483, 59, /* 0.23 */ }, + { 7009, 58, /* 0.23 */ }, + { 7558, 57, /* 0.22 */ }, + { 8130, 56, /* 0.22 */ }, + { 8726, 55, /* 0.22 */ }, + { 9346, 54, /* 0.21 */ }, + { 9991, 53, /* 0.21 */ }, + { 10661, 52, /* 0.21 */ }, + { 11358, 52, /* 0.20 */ }, + { 12082, 51, /* 0.20 */ }, + { 12834, 50, /* 0.20 */ }, + { 13614, 49, /* 0.19 */ }, + { 14424, 48, /* 0.19 */ }, + { 15265, 48, /* 0.19 */ }, + { 16137, 47, /* 0.19 */ }, + { 17042, 46, /* 0.18 */ }, + { 17981, 45, /* 0.18 */ }, + { 18955, 45, /* 0.18 */ }, + { 19965, 44, /* 0.17 */ }, + { 21013, 43, /* 0.17 */ }, + { 22101, 43, /* 0.17 */ }, + { 23230, 42, /* 0.17 */ }, + { 24402, 41, /* 0.16 */ }, + { 25618, 41, /* 0.16 */ }, + { 26881, 40, /* 0.16 */ }, + { 28193, 39, /* 0.16 */ }, + { 29557, 39, /* 0.15 */ }, + { 30975, 38, /* 0.15 */ }, + { 32450, 38, /* 0.15 */ }, + { 33986, 37, /* 0.15 */ }, + { 35586, 36, /* 0.14 */ }, + { 37253, 36, /* 0.14 */ }, + { 38992, 35, /* 0.14 */ }, + { 40808, 35, /* 0.14 */ }, + { 42707, 34, /* 0.13 */ }, + { 44694, 33, /* 0.13 */ }, + { 46776, 33, /* 0.13 */ }, + { 48961, 32, /* 0.13 */ }, + { 51258, 32, /* 0.13 */ }, + { 53677, 31, /* 0.12 */ }, + { 56230, 30, /* 0.12 */ }, + { 58932, 30, /* 0.12 */ }, + { 61799, 29, /* 0.12 */ }, + { 64851, 28, /* 0.11 */ }, + { 68113, 28, /* 0.11 */ }, + { 71617, 27, /* 0.11 */ }, + { 75401, 26, /* 0.10 */ }, + { 79517, 26, /* 0.10 */ }, + { 84035, 25, /* 0.10 */ }, + { 89053, 24, /* 0.10 */ }, +}; + +#define HSTCP_AIMD_MAX ARRAY_SIZE(hstcp_aimd_vals) + +/* Multiply the average RTT by ALPHA. The recommended value for + * ALPHA is 1.65. We approximate that. */ + +#define ALPHA_MULT_SHIFTS + +#ifdef ALPHA_MULT_SHIFTS +#define ALPHA_MULT(artt) \ + (artt + /* 1.00 = 1.00 */ \ + (artt >> 1) + /* + 0.50 = 1.50 */ \ + (artt >> 3) + /* + 0.125 = 1.625 */ \ + (artt >> 6) + /* + 0.015625 = 1.640625 */ \ + (artt >> 7)) /* + 0.0078125 = 1.6484375 */ +#else +#define ALPHA_MULT(artt) + ((artt * 33) / 20) /* 33/20 = 1.65 */ +#endif + +struct africa { + u32 minrtt; /* the minimum of all RTT measurements seen (in usec) */ + u32 artt; /* smoothed average of all RTT measurements seen (in usec) */ + u32 ai; /* used in the HSTCP fast mode */ + u8 fast_mode; /* boolean value specifying if we are in fast or slow mode */ +}; + +static void tcp_africa_reset(struct africa *ca) +{ + ca->minrtt = 0; + ca->artt = 0; + ca->fast_mode = 0; +} + +static void tcp_africa_init(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct africa *ca = inet_csk_ca(sk); + + ca->ai = 0; + + /* Ensure the MD arithmetic works. This is somewhat pedantic, + * since I don't think we will see a cwnd this large. :) */ + tp->snd_cwnd_clamp = min_t(u32, tp->snd_cwnd_clamp, 0xffffffff/128); + + tcp_africa_reset(ca); +} + +static void tcp_africa_rtt_calc(struct sock *sk, u32 usrtt) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct africa *ca = inet_csk_ca(sk); + u32 rtt, artt, minrtt; + + rtt = usrtt + 1; /* Never allow zero rtt */ + + minrtt = ca->minrtt; + artt = ca->artt; + + /* Get the smallest RTT seen. This is the assumed propogation delay. */ + if (rtt < minrtt || minrtt == 0) + minrtt = rtt; + + /* Get the average RTT so far. This is tha assumed propogation delay + * + queueing delay. + */ + if (artt > 0) + /* Find the exponentially smoothed average RTT. + * Using avg RTT = 7/8 avg RTT + 1/8 new RTT + * for speed, like in tcp_rtt_estimator. + */ + artt += (rtt >> 3) - (artt >> 3); + else + /* We don't have an average yet. */ + artt = rtt; + + /* Check how many packets are being queued in the network, and decide + * to do fast or slow mode. + */ + ca->fast_mode = (tp->snd_cwnd * (artt - minrtt) < ALPHA_MULT(artt)); + + ca->artt = artt; + ca->minrtt = minrtt; +} + +static void tcp_africa_cong_avoid(struct sock *sk, u32 ack, + u32 seq_rtt, u32 in_flight, int flag) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct africa *ca = inet_csk_ca(sk); + + if (!tcp_is_cwnd_limited(sk, in_flight)) + return; + + /* Adjust the cwnd */ + if (tp->snd_cwnd <= tp->snd_ssthresh) { + /* Standard TCP slow start. */ + tcp_slow_start(tp); + } + else if (ca->fast_mode) { + /* Fast increase (W = W + fast_increase(W)/W) + * + * Fast increase is done with the HSTCP algorithm. + */ + + /* Update AIMD parameters. + * + * We want to guarantee that: + * hstcp_aimd_vals[ca->ai-1].cwnd < + * snd_cwnd <= + * hstcp_aimd_vals[ca->ai].cwnd + */ + if (tp->snd_cwnd > hstcp_aimd_vals[ca->ai].cwnd) { + while (tp->snd_cwnd > + hstcp_aimd_vals[ca->ai].cwnd && + ca->ai < HSTCP_AIMD_MAX - 1) + ca->ai++; + } else if (ca->ai && tp->snd_cwnd <= + hstcp_aimd_vals[ca->ai-1].cwnd) { + while (ca->ai && tp->snd_cwnd <= + hstcp_aimd_vals[ca->ai-1].cwnd) + ca->ai--; + } + + /* Do additive increase */ + if (tp->snd_cwnd < tp->snd_cwnd_clamp) { + /* cwnd = cwnd + a(w) / cwnd */ + tp->snd_cwnd_cnt += ca->ai + 1; + if (tp->snd_cwnd_cnt >= tp->snd_cwnd) { + tp->snd_cwnd_cnt -= tp->snd_cwnd; + tp->snd_cwnd++; + } + } + } + else { + /* Slow increase (W = W + 1/W) + * + * Slow increase is done with the Reno algorithm. + */ + tcp_reno_cong_avoid(sk, ack, seq_rtt, in_flight, flag); + } +} + +static u32 tcp_africa_ssthresh(struct sock *sk) +{ + const struct tcp_sock *tp = tcp_sk(sk); + const struct africa *ca = inet_csk_ca(sk); + + /* Do multiplicative decrease */ + if (ca->fast_mode) { + /* HSTCP decrease. */ + return max(tp->snd_cwnd - + ((tp->snd_cwnd * hstcp_aimd_vals[ca->ai].md) >> 8), + 2U); + } + else { + /* Standard TCP Reno decrease. */ + return max(tp->snd_cwnd >> 1U, 2U); + } +} + +static struct tcp_congestion_ops tcp_africa = { + .init = tcp_africa_init, + .ssthresh = tcp_africa_ssthresh, + .cong_avoid = tcp_africa_cong_avoid, + .min_cwnd = tcp_reno_min_cwnd, + .rtt_sample = tcp_africa_rtt_calc, + + .owner = THIS_MODULE, + .name = "africa", +}; + +static int __init tcp_africa_register(void) +{ + BUILD_BUG_ON(sizeof(struct africa) > ICSK_CA_PRIV_SIZE); + tcp_register_congestion_control(&tcp_africa); + return 0; +} + +static void __exit tcp_africa_unregister(void) +{ + tcp_unregister_congestion_control(&tcp_africa); +} + +module_init(tcp_africa_register); +module_exit(tcp_africa_unregister); + +MODULE_AUTHOR("Dana Jansens"); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("TCP Africa");