linux/net/ipv4/tcp_cubic.c

558 lines
16 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* TCP CUBIC: Binary Increase Congestion control for TCP v2.3
* Home page:
* http://netsrv.csc.ncsu.edu/twiki/bin/view/Main/BIC
* This is from the implementation of CUBIC TCP in
* Sangtae Ha, Injong Rhee and Lisong Xu,
* "CUBIC: A New TCP-Friendly High-Speed TCP Variant"
* in ACM SIGOPS Operating System Review, July 2008.
* Available from:
* http://netsrv.csc.ncsu.edu/export/cubic_a_new_tcp_2008.pdf
*
* CUBIC integrates a new slow start algorithm, called HyStart.
* The details of HyStart are presented in
* Sangtae Ha and Injong Rhee,
* "Taming the Elephants: New TCP Slow Start", NCSU TechReport 2008.
* Available from:
* http://netsrv.csc.ncsu.edu/export/hystart_techreport_2008.pdf
*
* All testing results are available from:
* http://netsrv.csc.ncsu.edu/wiki/index.php/TCP_Testing
*
* Unless CUBIC is enabled and congestion window is large
* this behaves the same as the original Reno.
*/
#include <linux/mm.h>
#include <linux/btf.h>
#include <linux/btf_ids.h>
#include <linux/module.h>
#include <linux/math64.h>
#include <net/tcp.h>
#define BICTCP_BETA_SCALE 1024 /* Scale factor beta calculation
* max_cwnd = snd_cwnd * beta
*/
#define BICTCP_HZ 10 /* BIC HZ 2^10 = 1024 */
/* Two methods of hybrid slow start */
#define HYSTART_ACK_TRAIN 0x1
#define HYSTART_DELAY 0x2
/* Number of delay samples for detecting the increase of delay */
#define HYSTART_MIN_SAMPLES 8
#define HYSTART_DELAY_MIN (4000U) /* 4 ms */
#define HYSTART_DELAY_MAX (16000U) /* 16 ms */
#define HYSTART_DELAY_THRESH(x) clamp(x, HYSTART_DELAY_MIN, HYSTART_DELAY_MAX)
static int fast_convergence __read_mostly = 1;
static int beta __read_mostly = 717; /* = 717/1024 (BICTCP_BETA_SCALE) */
static int initial_ssthresh __read_mostly;
static int bic_scale __read_mostly = 41;
static int tcp_friendliness __read_mostly = 1;
static int hystart __read_mostly = 1;
static int hystart_detect __read_mostly = HYSTART_ACK_TRAIN | HYSTART_DELAY;
static int hystart_low_window __read_mostly = 16;
static int hystart_ack_delta_us __read_mostly = 2000;
static u32 cube_rtt_scale __read_mostly;
static u32 beta_scale __read_mostly;
static u64 cube_factor __read_mostly;
/* Note parameters that are used for precomputing scale factors are read-only */
module_param(fast_convergence, int, 0644);
MODULE_PARM_DESC(fast_convergence, "turn on/off fast convergence");
module_param(beta, int, 0644);
MODULE_PARM_DESC(beta, "beta for multiplicative increase");
module_param(initial_ssthresh, int, 0644);
MODULE_PARM_DESC(initial_ssthresh, "initial value of slow start threshold");
module_param(bic_scale, int, 0444);
MODULE_PARM_DESC(bic_scale, "scale (scaled by 1024) value for bic function (bic_scale/1024)");
module_param(tcp_friendliness, int, 0644);
MODULE_PARM_DESC(tcp_friendliness, "turn on/off tcp friendliness");
module_param(hystart, int, 0644);
MODULE_PARM_DESC(hystart, "turn on/off hybrid slow start algorithm");
module_param(hystart_detect, int, 0644);
MODULE_PARM_DESC(hystart_detect, "hybrid slow start detection mechanisms"
" 1: packet-train 2: delay 3: both packet-train and delay");
module_param(hystart_low_window, int, 0644);
MODULE_PARM_DESC(hystart_low_window, "lower bound cwnd for hybrid slow start");
module_param(hystart_ack_delta_us, int, 0644);
MODULE_PARM_DESC(hystart_ack_delta_us, "spacing between ack's indicating train (usecs)");
/* BIC TCP Parameters */
struct bictcp {
u32 cnt; /* increase cwnd by 1 after ACKs */
u32 last_max_cwnd; /* last maximum snd_cwnd */
u32 last_cwnd; /* the last snd_cwnd */
u32 last_time; /* time when updated last_cwnd */
u32 bic_origin_point;/* origin point of bic function */
u32 bic_K; /* time to origin point
from the beginning of the current epoch */
u32 delay_min; /* min delay (usec) */
u32 epoch_start; /* beginning of an epoch */
u32 ack_cnt; /* number of acks */
u32 tcp_cwnd; /* estimated tcp cwnd */
u16 unused;
u8 sample_cnt; /* number of samples to decide curr_rtt */
u8 found; /* the exit point is found? */
u32 round_start; /* beginning of each round */
u32 end_seq; /* end_seq of the round */
u32 last_ack; /* last time when the ACK spacing is close */
u32 curr_rtt; /* the minimum rtt of current round */
};
static inline void bictcp_reset(struct bictcp *ca)
{
memset(ca, 0, offsetof(struct bictcp, unused));
ca->found = 0;
}
static inline u32 bictcp_clock_us(const struct sock *sk)
{
return tcp_sk(sk)->tcp_mstamp;
}
static inline void bictcp_hystart_reset(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
struct bictcp *ca = inet_csk_ca(sk);
ca->round_start = ca->last_ack = bictcp_clock_us(sk);
ca->end_seq = tp->snd_nxt;
ca->curr_rtt = ~0U;
ca->sample_cnt = 0;
}
__bpf_kfunc static void cubictcp_init(struct sock *sk)
{
struct bictcp *ca = inet_csk_ca(sk);
bictcp_reset(ca);
if (hystart)
bictcp_hystart_reset(sk);
if (!hystart && initial_ssthresh)
tcp_sk(sk)->snd_ssthresh = initial_ssthresh;
}
__bpf_kfunc static void cubictcp_cwnd_event(struct sock *sk, enum tcp_ca_event event)
{
if (event == CA_EVENT_TX_START) {
struct bictcp *ca = inet_csk_ca(sk);
u32 now = tcp_jiffies32;
s32 delta;
delta = now - tcp_sk(sk)->lsndtime;
/* We were application limited (idle) for a while.
* Shift epoch_start to keep cwnd growth to cubic curve.
*/
if (ca->epoch_start && delta > 0) {
ca->epoch_start += delta;
if (after(ca->epoch_start, now))
ca->epoch_start = now;
}
return;
}
}
/* calculate the cubic root of x using a table lookup followed by one
* Newton-Raphson iteration.
* Avg err ~= 0.195%
*/
static u32 cubic_root(u64 a)
{
u32 x, b, shift;
/*
* cbrt(x) MSB values for x MSB values in [0..63].
* Precomputed then refined by hand - Willy Tarreau
*
* For x in [0..63],
* v = cbrt(x << 18) - 1
* cbrt(x) = (v[x] + 10) >> 6
*/
static const u8 v[] = {
/* 0x00 */ 0, 54, 54, 54, 118, 118, 118, 118,
/* 0x08 */ 123, 129, 134, 138, 143, 147, 151, 156,
/* 0x10 */ 157, 161, 164, 168, 170, 173, 176, 179,
/* 0x18 */ 181, 185, 187, 190, 192, 194, 197, 199,
/* 0x20 */ 200, 202, 204, 206, 209, 211, 213, 215,
/* 0x28 */ 217, 219, 221, 222, 224, 225, 227, 229,
/* 0x30 */ 231, 232, 234, 236, 237, 239, 240, 242,
/* 0x38 */ 244, 245, 246, 248, 250, 251, 252, 254,
};
b = fls64(a);
if (b < 7) {
/* a in [0..63] */
return ((u32)v[(u32)a] + 35) >> 6;
}
b = ((b * 84) >> 8) - 1;
shift = (a >> (b * 3));
x = ((u32)(((u32)v[shift] + 10) << b)) >> 6;
/*
* Newton-Raphson iteration
* 2
* x = ( 2 * x + a / x ) / 3
* k+1 k k
*/
x = (2 * x + (u32)div64_u64(a, (u64)x * (u64)(x - 1)));
x = ((x * 341) >> 10);
return x;
}
/*
* Compute congestion window to use.
*/
static inline void bictcp_update(struct bictcp *ca, u32 cwnd, u32 acked)
{
u32 delta, bic_target, max_cnt;
u64 offs, t;
ca->ack_cnt += acked; /* count the number of ACKed packets */
if (ca->last_cwnd == cwnd &&
(s32)(tcp_jiffies32 - ca->last_time) <= HZ / 32)
return;
tcp: fix timing issue in CUBIC slope calculation This patch fixes a bug in CUBIC that causes cwnd to increase slightly too slowly when multiple ACKs arrive in the same jiffy. If cwnd is supposed to increase at a rate of more than once per jiffy, then CUBIC was sometimes too slow. Because the bic_target is calculated for a future point in time, calculated with time in jiffies, the cwnd can increase over the course of the jiffy while the bic_target calculated as the proper CUBIC cwnd at time t=tcp_time_stamp+rtt does not increase, because tcp_time_stamp only increases on jiffy tick boundaries. So since the cnt is set to: ca->cnt = cwnd / (bic_target - cwnd); as cwnd increases but bic_target does not increase due to jiffy granularity, the cnt becomes too large, causing cwnd to increase too slowly. For example: - suppose at the beginning of a jiffy, cwnd=40, bic_target=44 - so CUBIC sets: ca->cnt = cwnd / (bic_target - cwnd) = 40 / (44 - 40) = 40/4 = 10 - suppose we get 10 acks, each for 1 segment, so tcp_cong_avoid_ai() increases cwnd to 41 - so CUBIC sets: ca->cnt = cwnd / (bic_target - cwnd) = 41 / (44 - 41) = 41 / 3 = 13 So now CUBIC will wait for 13 packets to be ACKed before increasing cwnd to 42, insted of 10 as it should. The fix is to avoid adjusting the slope (determined by ca->cnt) multiple times within a jiffy, and instead skip to compute the Reno cwnd, the "TCP friendliness" code path. Reported-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-29 09:01:39 +08:00
/* The CUBIC function can update ca->cnt at most once per jiffy.
* On all cwnd reduction events, ca->epoch_start is set to 0,
* which will force a recalculation of ca->cnt.
*/
if (ca->epoch_start && tcp_jiffies32 == ca->last_time)
tcp: fix timing issue in CUBIC slope calculation This patch fixes a bug in CUBIC that causes cwnd to increase slightly too slowly when multiple ACKs arrive in the same jiffy. If cwnd is supposed to increase at a rate of more than once per jiffy, then CUBIC was sometimes too slow. Because the bic_target is calculated for a future point in time, calculated with time in jiffies, the cwnd can increase over the course of the jiffy while the bic_target calculated as the proper CUBIC cwnd at time t=tcp_time_stamp+rtt does not increase, because tcp_time_stamp only increases on jiffy tick boundaries. So since the cnt is set to: ca->cnt = cwnd / (bic_target - cwnd); as cwnd increases but bic_target does not increase due to jiffy granularity, the cnt becomes too large, causing cwnd to increase too slowly. For example: - suppose at the beginning of a jiffy, cwnd=40, bic_target=44 - so CUBIC sets: ca->cnt = cwnd / (bic_target - cwnd) = 40 / (44 - 40) = 40/4 = 10 - suppose we get 10 acks, each for 1 segment, so tcp_cong_avoid_ai() increases cwnd to 41 - so CUBIC sets: ca->cnt = cwnd / (bic_target - cwnd) = 41 / (44 - 41) = 41 / 3 = 13 So now CUBIC will wait for 13 packets to be ACKed before increasing cwnd to 42, insted of 10 as it should. The fix is to avoid adjusting the slope (determined by ca->cnt) multiple times within a jiffy, and instead skip to compute the Reno cwnd, the "TCP friendliness" code path. Reported-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-29 09:01:39 +08:00
goto tcp_friendliness;
ca->last_cwnd = cwnd;
ca->last_time = tcp_jiffies32;
if (ca->epoch_start == 0) {
ca->epoch_start = tcp_jiffies32; /* record beginning */
ca->ack_cnt = acked; /* start counting */
ca->tcp_cwnd = cwnd; /* syn with cubic */
if (ca->last_max_cwnd <= cwnd) {
ca->bic_K = 0;
ca->bic_origin_point = cwnd;
} else {
/* Compute new K based on
* (wmax-cwnd) * (srtt>>3 / HZ) / c * 2^(3*bictcp_HZ)
*/
ca->bic_K = cubic_root(cube_factor
* (ca->last_max_cwnd - cwnd));
ca->bic_origin_point = ca->last_max_cwnd;
}
}
/* cubic function - calc*/
/* calculate c * time^3 / rtt,
* while considering overflow in calculation of time^3
* (so time^3 is done by using 64 bit)
* and without the support of division of 64bit numbers
* (so all divisions are done by using 32 bit)
* also NOTE the unit of those veriables
* time = (t - K) / 2^bictcp_HZ
* c = bic_scale >> 10
* rtt = (srtt >> 3) / HZ
* !!! The following code does not have overflow problems,
* if the cwnd < 1 million packets !!!
*/
t = (s32)(tcp_jiffies32 - ca->epoch_start);
t += usecs_to_jiffies(ca->delay_min);
/* change the unit from HZ to bictcp_HZ */
t <<= BICTCP_HZ;
do_div(t, HZ);
if (t < ca->bic_K) /* t - K */
offs = ca->bic_K - t;
else
offs = t - ca->bic_K;
/* c/rtt * (t-K)^3 */
delta = (cube_rtt_scale * offs * offs * offs) >> (10+3*BICTCP_HZ);
if (t < ca->bic_K) /* below origin*/
bic_target = ca->bic_origin_point - delta;
else /* above origin*/
bic_target = ca->bic_origin_point + delta;
/* cubic function - calc bictcp_cnt*/
if (bic_target > cwnd) {
ca->cnt = cwnd / (bic_target - cwnd);
} else {
ca->cnt = 100 * cwnd; /* very small increment*/
}
/*
* The initial growth of cubic function may be too conservative
* when the available bandwidth is still unknown.
*/
if (ca->last_max_cwnd == 0 && ca->cnt > 20)
ca->cnt = 20; /* increase cwnd 5% per RTT */
tcp: fix timing issue in CUBIC slope calculation This patch fixes a bug in CUBIC that causes cwnd to increase slightly too slowly when multiple ACKs arrive in the same jiffy. If cwnd is supposed to increase at a rate of more than once per jiffy, then CUBIC was sometimes too slow. Because the bic_target is calculated for a future point in time, calculated with time in jiffies, the cwnd can increase over the course of the jiffy while the bic_target calculated as the proper CUBIC cwnd at time t=tcp_time_stamp+rtt does not increase, because tcp_time_stamp only increases on jiffy tick boundaries. So since the cnt is set to: ca->cnt = cwnd / (bic_target - cwnd); as cwnd increases but bic_target does not increase due to jiffy granularity, the cnt becomes too large, causing cwnd to increase too slowly. For example: - suppose at the beginning of a jiffy, cwnd=40, bic_target=44 - so CUBIC sets: ca->cnt = cwnd / (bic_target - cwnd) = 40 / (44 - 40) = 40/4 = 10 - suppose we get 10 acks, each for 1 segment, so tcp_cong_avoid_ai() increases cwnd to 41 - so CUBIC sets: ca->cnt = cwnd / (bic_target - cwnd) = 41 / (44 - 41) = 41 / 3 = 13 So now CUBIC will wait for 13 packets to be ACKed before increasing cwnd to 42, insted of 10 as it should. The fix is to avoid adjusting the slope (determined by ca->cnt) multiple times within a jiffy, and instead skip to compute the Reno cwnd, the "TCP friendliness" code path. Reported-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-29 09:01:39 +08:00
tcp_friendliness:
/* TCP Friendly */
if (tcp_friendliness) {
u32 scale = beta_scale;
delta = (cwnd * scale) >> 3;
while (ca->ack_cnt > delta) { /* update tcp cwnd */
ca->ack_cnt -= delta;
ca->tcp_cwnd++;
}
if (ca->tcp_cwnd > cwnd) { /* if bic is slower than tcp */
delta = ca->tcp_cwnd - cwnd;
max_cnt = cwnd / delta;
if (ca->cnt > max_cnt)
ca->cnt = max_cnt;
}
}
/* The maximum rate of cwnd increase CUBIC allows is 1 packet per
* 2 packets ACKed, meaning cwnd grows at 1.5x per RTT.
*/
ca->cnt = max(ca->cnt, 2U);
}
__bpf_kfunc static void cubictcp_cong_avoid(struct sock *sk, u32 ack, u32 acked)
{
struct tcp_sock *tp = tcp_sk(sk);
struct bictcp *ca = inet_csk_ca(sk);
if (!tcp_is_cwnd_limited(sk))
return;
if (tcp_in_slow_start(tp)) {
acked = tcp_slow_start(tp, acked);
if (!acked)
return;
}
bictcp_update(ca, tcp_snd_cwnd(tp), acked);
tcp_cong_avoid_ai(tp, ca->cnt, acked);
}
__bpf_kfunc static u32 cubictcp_recalc_ssthresh(struct sock *sk)
{
const struct tcp_sock *tp = tcp_sk(sk);
struct bictcp *ca = inet_csk_ca(sk);
ca->epoch_start = 0; /* end of epoch */
/* Wmax and fast convergence */
if (tcp_snd_cwnd(tp) < ca->last_max_cwnd && fast_convergence)
ca->last_max_cwnd = (tcp_snd_cwnd(tp) * (BICTCP_BETA_SCALE + beta))
/ (2 * BICTCP_BETA_SCALE);
else
ca->last_max_cwnd = tcp_snd_cwnd(tp);
return max((tcp_snd_cwnd(tp) * beta) / BICTCP_BETA_SCALE, 2U);
}
__bpf_kfunc static void cubictcp_state(struct sock *sk, u8 new_state)
{
if (new_state == TCP_CA_Loss) {
bictcp_reset(inet_csk_ca(sk));
bictcp_hystart_reset(sk);
}
}
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
/* Account for TSO/GRO delays.
* Otherwise short RTT flows could get too small ssthresh, since during
* slow start we begin with small TSO packets and ca->delay_min would
* not account for long aggregation delay when TSO packets get bigger.
* Ideally even with a very small RTT we would like to have at least one
* TSO packet being sent and received by GRO, and another one in qdisc layer.
* We apply another 100% factor because @rate is doubled at this point.
* We cap the cushion to 1ms.
*/
static u32 hystart_ack_delay(const struct sock *sk)
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
{
unsigned long rate;
rate = READ_ONCE(sk->sk_pacing_rate);
if (!rate)
return 0;
return min_t(u64, USEC_PER_MSEC,
div64_ul((u64)sk->sk_gso_max_size * 4 * USEC_PER_SEC, rate));
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
}
static void hystart_update(struct sock *sk, u32 delay)
{
struct tcp_sock *tp = tcp_sk(sk);
struct bictcp *ca = inet_csk_ca(sk);
u32 threshold;
tcp_cubic: fix spurious Hystart ACK train detections for not-cwnd-limited flows While testing BIG TCP patch series, I was expecting that TCP_RR workloads with 80KB requests/answers would send one 80KB TSO packet, then being received as a single GRO packet. It turns out this was not happening, and the root cause was that cubic Hystart ACK train was triggering after a few (2 or 3) rounds of RPC. Hystart was wrongly setting CWND/SSTHRESH to 30, while my RPC needed a budget of ~20 segments. Ideally these TCP_RR flows should not exit slow start. Cubic Hystart should reset itself at each round, instead of assuming every TCP flow is a bulk one. Note that even after this patch, Hystart can still trigger, depending on scheduling artifacts, but at a higher CWND/SSTHRESH threshold, keeping optimal TSO packet sizes. Tested: ip link set dev eth0 gro_ipv6_max_size 131072 gso_ipv6_max_size 131072 nstat -n; netperf -H ... -t TCP_RR -l 5 -- -r 80000,80000 -K cubic; nstat|egrep "Ip6InReceives|Hystart|Ip6OutRequests" Before: 8605 Ip6InReceives 87541 0.0 Ip6OutRequests 129496 0.0 TcpExtTCPHystartTrainDetect 1 0.0 TcpExtTCPHystartTrainCwnd 30 0.0 After: 8760 Ip6InReceives 88514 0.0 Ip6OutRequests 87975 0.0 Fixes: ae27e98a5152 ("[TCP] CUBIC v2.3") Co-developed-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Stephen Hemminger <stephen@networkplumber.org> Cc: Yuchung Cheng <ycheng@google.com> Cc: Soheil Hassas Yeganeh <soheil@google.com> Link: https://lore.kernel.org/r/20211123202535.1843771-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2021-11-24 04:25:35 +08:00
if (after(tp->snd_una, ca->end_seq))
bictcp_hystart_reset(sk);
if (hystart_detect & HYSTART_ACK_TRAIN) {
u32 now = bictcp_clock_us(sk);
/* first detection parameter - ack-train detection */
if ((s32)(now - ca->last_ack) <= hystart_ack_delta_us) {
ca->last_ack = now;
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
threshold = ca->delay_min + hystart_ack_delay(sk);
/* Hystart ack train triggers if we get ack past
* ca->delay_min/2.
* Pacing might have delayed packets up to RTT/2
* during slow start.
*/
if (sk->sk_pacing_status == SK_PACING_NONE)
threshold >>= 1;
if ((s32)(now - ca->round_start) > threshold) {
ca->found = 1;
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
pr_debug("hystart_ack_train (%u > %u) delay_min %u (+ ack_delay %u) cwnd %u\n",
now - ca->round_start, threshold,
ca->delay_min, hystart_ack_delay(sk), tcp_snd_cwnd(tp));
NET_INC_STATS(sock_net(sk),
LINUX_MIB_TCPHYSTARTTRAINDETECT);
NET_ADD_STATS(sock_net(sk),
LINUX_MIB_TCPHYSTARTTRAINCWND,
tcp_snd_cwnd(tp));
tp->snd_ssthresh = tcp_snd_cwnd(tp);
}
}
}
if (hystart_detect & HYSTART_DELAY) {
/* obtain the minimum delay of more than sampling packets */
if (ca->curr_rtt > delay)
ca->curr_rtt = delay;
if (ca->sample_cnt < HYSTART_MIN_SAMPLES) {
ca->sample_cnt++;
} else {
if (ca->curr_rtt > ca->delay_min +
HYSTART_DELAY_THRESH(ca->delay_min >> 3)) {
ca->found = 1;
NET_INC_STATS(sock_net(sk),
LINUX_MIB_TCPHYSTARTDELAYDETECT);
NET_ADD_STATS(sock_net(sk),
LINUX_MIB_TCPHYSTARTDELAYCWND,
tcp_snd_cwnd(tp));
tp->snd_ssthresh = tcp_snd_cwnd(tp);
}
}
}
}
__bpf_kfunc static void cubictcp_acked(struct sock *sk, const struct ack_sample *sample)
{
const struct tcp_sock *tp = tcp_sk(sk);
struct bictcp *ca = inet_csk_ca(sk);
u32 delay;
/* Some calls are for duplicates without timetamps */
if (sample->rtt_us < 0)
return;
/* Discard delay samples right after fast recovery */
if (ca->epoch_start && (s32)(tcp_jiffies32 - ca->epoch_start) < HZ)
return;
delay = sample->rtt_us;
if (delay == 0)
delay = 1;
/* first time call or link delay decreases */
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
if (ca->delay_min == 0 || ca->delay_min > delay)
ca->delay_min = delay;
/* hystart triggers when cwnd is larger than some threshold */
tcp_cubic: refactor code to perform a divide only when needed Neal Cardwell suggested to not change ca->delay_min and apply the ack delay cushion only when Hystart ACK train is still under consideration. This should avoid a 64bit divide unless needed. Tested: 40Gbit(mlx4) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpaa24 -l -4000000; done;nstat|egrep "Hystart" 14815 16280 15293 15563 11574 15145 14789 18548 16972 12520 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 1396 0.0 $ dmesg | tail -10 [ 4873.951350] hystart_ack_train (116 > 93) delay_min 24 (+ ack_delay 69) cwnd 80 [ 4875.155379] hystart_ack_train (55 > 50) delay_min 21 (+ ack_delay 29) cwnd 160 [ 4876.333921] hystart_ack_train (69 > 62) delay_min 23 (+ ack_delay 39) cwnd 130 [ 4877.519037] hystart_ack_train (69 > 60) delay_min 22 (+ ack_delay 38) cwnd 130 [ 4878.701559] hystart_ack_train (87 > 63) delay_min 24 (+ ack_delay 39) cwnd 160 [ 4879.844597] hystart_ack_train (93 > 50) delay_min 21 (+ ack_delay 29) cwnd 216 [ 4880.956650] hystart_ack_train (74 > 67) delay_min 20 (+ ack_delay 47) cwnd 108 [ 4882.098500] hystart_ack_train (61 > 57) delay_min 23 (+ ack_delay 34) cwnd 130 [ 4883.262056] hystart_ack_train (72 > 67) delay_min 21 (+ ack_delay 46) cwnd 130 [ 4884.418760] hystart_ack_train (74 > 67) delay_min 29 (+ ack_delay 38) cwnd 152 10Gbit(bnx2x) testbed (with sch_fq as packet scheduler) $ echo -n 'file tcp_cubic.c +p' >/sys/kernel/debug/dynamic_debug/control $ nstat -n;for f in {1..10}; do ./super_netperf 1 -H lpk52 -l -4000000; done;nstat|egrep "Hystart" 7050 7065 7100 6900 7202 7263 7189 6869 7463 7034 TcpExtTCPHystartTrainDetect 10 0.0 TcpExtTCPHystartTrainCwnd 3199 0.0 $ dmesg | tail -10 [ 176.920012] hystart_ack_train (161 > 141) delay_min 83 (+ ack_delay 58) cwnd 264 [ 179.144645] hystart_ack_train (164 > 159) delay_min 120 (+ ack_delay 39) cwnd 444 [ 181.354527] hystart_ack_train (214 > 168) delay_min 125 (+ ack_delay 43) cwnd 436 [ 183.539565] hystart_ack_train (170 > 147) delay_min 96 (+ ack_delay 51) cwnd 326 [ 185.727309] hystart_ack_train (177 > 160) delay_min 61 (+ ack_delay 99) cwnd 128 [ 187.947142] hystart_ack_train (184 > 167) delay_min 123 (+ ack_delay 44) cwnd 367 [ 190.166680] hystart_ack_train (230 > 153) delay_min 116 (+ ack_delay 37) cwnd 444 [ 192.327285] hystart_ack_train (210 > 206) delay_min 86 (+ ack_delay 120) cwnd 152 [ 194.511392] hystart_ack_train (173 > 151) delay_min 94 (+ ack_delay 57) cwnd 239 [ 196.736023] hystart_ack_train (149 > 146) delay_min 105 (+ ack_delay 41) cwnd 399 Fixes: 42f3a8aaae66 ("tcp_cubic: tweak Hystart detection for short RTT flows") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Neal Cardwell <ncardwell@google.com> Link: https://www.spinics.net/lists/netdev/msg621886.html Link: https://www.spinics.net/lists/netdev/msg621797.html Acked-by: Neal Cardwell <ncardwell@google.com> Acked-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-12-30 22:06:19 +08:00
if (!ca->found && tcp_in_slow_start(tp) && hystart &&
tcp_snd_cwnd(tp) >= hystart_low_window)
hystart_update(sk, delay);
}
static struct tcp_congestion_ops cubictcp __read_mostly = {
.init = cubictcp_init,
.ssthresh = cubictcp_recalc_ssthresh,
.cong_avoid = cubictcp_cong_avoid,
.set_state = cubictcp_state,
.undo_cwnd = tcp_reno_undo_cwnd,
.cwnd_event = cubictcp_cwnd_event,
.pkts_acked = cubictcp_acked,
.owner = THIS_MODULE,
.name = "cubic",
};
BTF_KFUNCS_START(tcp_cubic_check_kfunc_ids)
#ifdef CONFIG_X86
#ifdef CONFIG_DYNAMIC_FTRACE
BTF_ID_FLAGS(func, cubictcp_init)
BTF_ID_FLAGS(func, cubictcp_recalc_ssthresh)
BTF_ID_FLAGS(func, cubictcp_cong_avoid)
BTF_ID_FLAGS(func, cubictcp_state)
BTF_ID_FLAGS(func, cubictcp_cwnd_event)
BTF_ID_FLAGS(func, cubictcp_acked)
#endif
#endif
BTF_KFUNCS_END(tcp_cubic_check_kfunc_ids)
static const struct btf_kfunc_id_set tcp_cubic_kfunc_set = {
.owner = THIS_MODULE,
.set = &tcp_cubic_check_kfunc_ids,
};
static int __init cubictcp_register(void)
{
int ret;
BUILD_BUG_ON(sizeof(struct bictcp) > ICSK_CA_PRIV_SIZE);
/* Precompute a bunch of the scaling factors that are used per-packet
* based on SRTT of 100ms
*/
beta_scale = 8*(BICTCP_BETA_SCALE+beta) / 3
/ (BICTCP_BETA_SCALE - beta);
cube_rtt_scale = (bic_scale * 10); /* 1024*c/rtt */
/* calculate the "K" for (wmax-cwnd) = c/rtt * K^3
* so K = cubic_root( (wmax-cwnd)*rtt/c )
* the unit of K is bictcp_HZ=2^10, not HZ
*
* c = bic_scale >> 10
* rtt = 100ms
*
* the following code has been designed and tested for
* cwnd < 1 million packets
* RTT < 100 seconds
* HZ < 1,000,00 (corresponding to 10 nano-second)
*/
/* 1/c * 2^2*bictcp_HZ * srtt */
cube_factor = 1ull << (10+3*BICTCP_HZ); /* 2^40 */
/* divide by bic_scale and by constant Srtt (100ms) */
do_div(cube_factor, bic_scale * 10);
ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &tcp_cubic_kfunc_set);
if (ret < 0)
return ret;
return tcp_register_congestion_control(&cubictcp);
}
static void __exit cubictcp_unregister(void)
{
tcp_unregister_congestion_control(&cubictcp);
}
module_init(cubictcp_register);
module_exit(cubictcp_unregister);
MODULE_AUTHOR("Sangtae Ha, Stephen Hemminger");
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("CUBIC TCP");
MODULE_VERSION("2.3");