1 // SPDX-License-Identifier: GPL-2.0-only
3 /* WARNING: This implemenation is not necessarily the same
4 * as the tcp_cubic.c. The purpose is mainly for testing
5 * the kernel BPF logic.
8 * 1. CONFIG_HZ .kconfig map is used.
9 * 2. In bictcp_update(), calculation is changed to use usec
10 * resolution (i.e. USEC_PER_JIFFY) instead of using jiffies.
11 * Thus, usecs_to_jiffies() is not used in the bpf_cubic.c.
12 * 3. In bitctcp_update() [under tcp_friendliness], the original
13 * "while (ca->ack_cnt > delta)" loop is changed to the equivalent
14 * "ca->ack_cnt / delta" operation.
17 #include <linux/bpf.h>
18 #include <linux/stddef.h>
19 #include <linux/tcp.h>
20 #include "bpf_tcp_helpers.h"
22 char _license[] SEC("license") = "GPL";
24 #define clamp(val, lo, hi) min((typeof(val))max(val, lo), hi)
26 #define BICTCP_BETA_SCALE 1024 /* Scale factor beta calculation
27 * max_cwnd = snd_cwnd * beta
29 #define BICTCP_HZ 10 /* BIC HZ 2^10 = 1024 */
31 /* Two methods of hybrid slow start */
32 #define HYSTART_ACK_TRAIN 0x1
33 #define HYSTART_DELAY 0x2
35 /* Number of delay samples for detecting the increase of delay */
36 #define HYSTART_MIN_SAMPLES 8
37 #define HYSTART_DELAY_MIN (4000U) /* 4ms */
38 #define HYSTART_DELAY_MAX (16000U) /* 16 ms */
39 #define HYSTART_DELAY_THRESH(x) clamp(x, HYSTART_DELAY_MIN, HYSTART_DELAY_MAX)
41 static int fast_convergence = 1;
42 static const int beta = 717; /* = 717/1024 (BICTCP_BETA_SCALE) */
43 static int initial_ssthresh;
44 static const int bic_scale = 41;
45 static int tcp_friendliness = 1;
47 static int hystart = 1;
48 static int hystart_detect = HYSTART_ACK_TRAIN | HYSTART_DELAY;
49 static int hystart_low_window = 16;
50 static int hystart_ack_delta_us = 2000;
52 static const __u32 cube_rtt_scale = (bic_scale * 10); /* 1024*c/rtt */
53 static const __u32 beta_scale = 8*(BICTCP_BETA_SCALE+beta) / 3
54 / (BICTCP_BETA_SCALE - beta);
55 /* calculate the "K" for (wmax-cwnd) = c/rtt * K^3
56 * so K = cubic_root( (wmax-cwnd)*rtt/c )
57 * the unit of K is bictcp_HZ=2^10, not HZ
62 * the following code has been designed and tested for
63 * cwnd < 1 million packets
65 * HZ < 1,000,00 (corresponding to 10 nano-second)
68 /* 1/c * 2^2*bictcp_HZ * srtt, 2^40 */
69 static const __u64 cube_factor = (__u64)(1ull << (10+3*BICTCP_HZ))
72 /* BIC TCP Parameters */
74 __u32 cnt; /* increase cwnd by 1 after ACKs */
75 __u32 last_max_cwnd; /* last maximum snd_cwnd */
76 __u32 last_cwnd; /* the last snd_cwnd */
77 __u32 last_time; /* time when updated last_cwnd */
78 __u32 bic_origin_point;/* origin point of bic function */
79 __u32 bic_K; /* time to origin point
80 from the beginning of the current epoch */
81 __u32 delay_min; /* min delay (usec) */
82 __u32 epoch_start; /* beginning of an epoch */
83 __u32 ack_cnt; /* number of acks */
84 __u32 tcp_cwnd; /* estimated tcp cwnd */
86 __u8 sample_cnt; /* number of samples to decide curr_rtt */
87 __u8 found; /* the exit point is found? */
88 __u32 round_start; /* beginning of each round */
89 __u32 end_seq; /* end_seq of the round */
90 __u32 last_ack; /* last time when the ACK spacing is close */
91 __u32 curr_rtt; /* the minimum rtt of current round */
94 static inline void bictcp_reset(struct bictcp *ca)
97 ca->last_max_cwnd = 0;
100 ca->bic_origin_point = 0;
109 extern unsigned long CONFIG_HZ __kconfig;
111 #define USEC_PER_MSEC 1000UL
112 #define USEC_PER_SEC 1000000UL
113 #define USEC_PER_JIFFY (USEC_PER_SEC / HZ)
115 static __always_inline __u64 div64_u64(__u64 dividend, __u64 divisor)
117 return dividend / divisor;
120 #define div64_ul div64_u64
122 #define BITS_PER_U64 (sizeof(__u64) * 8)
123 static __always_inline int fls64(__u64 x)
125 int num = BITS_PER_U64 - 1;
130 if (!(x & (~0ull << (BITS_PER_U64-32)))) {
134 if (!(x & (~0ull << (BITS_PER_U64-16)))) {
138 if (!(x & (~0ull << (BITS_PER_U64-8)))) {
142 if (!(x & (~0ull << (BITS_PER_U64-4)))) {
146 if (!(x & (~0ull << (BITS_PER_U64-2)))) {
150 if (!(x & (~0ull << (BITS_PER_U64-1))))
156 static __always_inline __u32 bictcp_clock_us(const struct sock *sk)
158 return tcp_sk(sk)->tcp_mstamp;
161 static __always_inline void bictcp_hystart_reset(struct sock *sk)
163 struct tcp_sock *tp = tcp_sk(sk);
164 struct bictcp *ca = inet_csk_ca(sk);
166 ca->round_start = ca->last_ack = bictcp_clock_us(sk);
167 ca->end_seq = tp->snd_nxt;
172 /* "struct_ops/" prefix is a requirement */
173 SEC("struct_ops/bpf_cubic_init")
174 void BPF_PROG(bpf_cubic_init, struct sock *sk)
176 struct bictcp *ca = inet_csk_ca(sk);
181 bictcp_hystart_reset(sk);
183 if (!hystart && initial_ssthresh)
184 tcp_sk(sk)->snd_ssthresh = initial_ssthresh;
187 /* "struct_ops" prefix is a requirement */
188 SEC("struct_ops/bpf_cubic_cwnd_event")
189 void BPF_PROG(bpf_cubic_cwnd_event, struct sock *sk, enum tcp_ca_event event)
191 if (event == CA_EVENT_TX_START) {
192 struct bictcp *ca = inet_csk_ca(sk);
193 __u32 now = tcp_jiffies32;
196 delta = now - tcp_sk(sk)->lsndtime;
198 /* We were application limited (idle) for a while.
199 * Shift epoch_start to keep cwnd growth to cubic curve.
201 if (ca->epoch_start && delta > 0) {
202 ca->epoch_start += delta;
203 if (after(ca->epoch_start, now))
204 ca->epoch_start = now;
211 * cbrt(x) MSB values for x MSB values in [0..63].
212 * Precomputed then refined by hand - Willy Tarreau
215 * v = cbrt(x << 18) - 1
216 * cbrt(x) = (v[x] + 10) >> 6
218 static const __u8 v[] = {
219 /* 0x00 */ 0, 54, 54, 54, 118, 118, 118, 118,
220 /* 0x08 */ 123, 129, 134, 138, 143, 147, 151, 156,
221 /* 0x10 */ 157, 161, 164, 168, 170, 173, 176, 179,
222 /* 0x18 */ 181, 185, 187, 190, 192, 194, 197, 199,
223 /* 0x20 */ 200, 202, 204, 206, 209, 211, 213, 215,
224 /* 0x28 */ 217, 219, 221, 222, 224, 225, 227, 229,
225 /* 0x30 */ 231, 232, 234, 236, 237, 239, 240, 242,
226 /* 0x38 */ 244, 245, 246, 248, 250, 251, 252, 254,
229 /* calculate the cubic root of x using a table lookup followed by one
230 * Newton-Raphson iteration.
233 static __always_inline __u32 cubic_root(__u64 a)
239 return ((__u32)v[(__u32)a] + 35) >> 6;
243 b = ((b * 84) >> 8) - 1;
244 shift = (a >> (b * 3));
246 /* it is needed for verifier's bound check on v */
250 x = ((__u32)(((__u32)v[shift] + 10) << b)) >> 6;
253 * Newton-Raphson iteration
255 * x = ( 2 * x + a / x ) / 3
258 x = (2 * x + (__u32)div64_u64(a, (__u64)x * (__u64)(x - 1)));
259 x = ((x * 341) >> 10);
264 * Compute congestion window to use.
266 static __always_inline void bictcp_update(struct bictcp *ca, __u32 cwnd,
269 __u32 delta, bic_target, max_cnt;
272 ca->ack_cnt += acked; /* count the number of ACKed packets */
274 if (ca->last_cwnd == cwnd &&
275 (__s32)(tcp_jiffies32 - ca->last_time) <= HZ / 32)
278 /* The CUBIC function can update ca->cnt at most once per jiffy.
279 * On all cwnd reduction events, ca->epoch_start is set to 0,
280 * which will force a recalculation of ca->cnt.
282 if (ca->epoch_start && tcp_jiffies32 == ca->last_time)
283 goto tcp_friendliness;
285 ca->last_cwnd = cwnd;
286 ca->last_time = tcp_jiffies32;
288 if (ca->epoch_start == 0) {
289 ca->epoch_start = tcp_jiffies32; /* record beginning */
290 ca->ack_cnt = acked; /* start counting */
291 ca->tcp_cwnd = cwnd; /* syn with cubic */
293 if (ca->last_max_cwnd <= cwnd) {
295 ca->bic_origin_point = cwnd;
297 /* Compute new K based on
298 * (wmax-cwnd) * (srtt>>3 / HZ) / c * 2^(3*bictcp_HZ)
300 ca->bic_K = cubic_root(cube_factor
301 * (ca->last_max_cwnd - cwnd));
302 ca->bic_origin_point = ca->last_max_cwnd;
306 /* cubic function - calc*/
307 /* calculate c * time^3 / rtt,
308 * while considering overflow in calculation of time^3
309 * (so time^3 is done by using 64 bit)
310 * and without the support of division of 64bit numbers
311 * (so all divisions are done by using 32 bit)
312 * also NOTE the unit of those veriables
313 * time = (t - K) / 2^bictcp_HZ
314 * c = bic_scale >> 10
315 * rtt = (srtt >> 3) / HZ
316 * !!! The following code does not have overflow problems,
317 * if the cwnd < 1 million packets !!!
320 t = (__s32)(tcp_jiffies32 - ca->epoch_start) * USEC_PER_JIFFY;
322 /* change the unit from usec to bictcp_HZ */
326 if (t < ca->bic_K) /* t - K */
327 offs = ca->bic_K - t;
329 offs = t - ca->bic_K;
331 /* c/rtt * (t-K)^3 */
332 delta = (cube_rtt_scale * offs * offs * offs) >> (10+3*BICTCP_HZ);
333 if (t < ca->bic_K) /* below origin*/
334 bic_target = ca->bic_origin_point - delta;
335 else /* above origin*/
336 bic_target = ca->bic_origin_point + delta;
338 /* cubic function - calc bictcp_cnt*/
339 if (bic_target > cwnd) {
340 ca->cnt = cwnd / (bic_target - cwnd);
342 ca->cnt = 100 * cwnd; /* very small increment*/
346 * The initial growth of cubic function may be too conservative
347 * when the available bandwidth is still unknown.
349 if (ca->last_max_cwnd == 0 && ca->cnt > 20)
350 ca->cnt = 20; /* increase cwnd 5% per RTT */
354 if (tcp_friendliness) {
355 __u32 scale = beta_scale;
358 /* update tcp cwnd */
359 delta = (cwnd * scale) >> 3;
360 if (ca->ack_cnt > delta && delta) {
361 n = ca->ack_cnt / delta;
362 ca->ack_cnt -= n * delta;
366 if (ca->tcp_cwnd > cwnd) { /* if bic is slower than tcp */
367 delta = ca->tcp_cwnd - cwnd;
368 max_cnt = cwnd / delta;
369 if (ca->cnt > max_cnt)
374 /* The maximum rate of cwnd increase CUBIC allows is 1 packet per
375 * 2 packets ACKed, meaning cwnd grows at 1.5x per RTT.
377 ca->cnt = max(ca->cnt, 2U);
380 /* Or simply use the BPF_STRUCT_OPS to avoid the SEC boiler plate. */
381 void BPF_STRUCT_OPS(bpf_cubic_cong_avoid, struct sock *sk, __u32 ack, __u32 acked)
383 struct tcp_sock *tp = tcp_sk(sk);
384 struct bictcp *ca = inet_csk_ca(sk);
386 if (!tcp_is_cwnd_limited(sk))
389 if (tcp_in_slow_start(tp)) {
390 if (hystart && after(ack, ca->end_seq))
391 bictcp_hystart_reset(sk);
392 acked = tcp_slow_start(tp, acked);
396 bictcp_update(ca, tp->snd_cwnd, acked);
397 tcp_cong_avoid_ai(tp, ca->cnt, acked);
400 __u32 BPF_STRUCT_OPS(bpf_cubic_recalc_ssthresh, struct sock *sk)
402 const struct tcp_sock *tp = tcp_sk(sk);
403 struct bictcp *ca = inet_csk_ca(sk);
405 ca->epoch_start = 0; /* end of epoch */
407 /* Wmax and fast convergence */
408 if (tp->snd_cwnd < ca->last_max_cwnd && fast_convergence)
409 ca->last_max_cwnd = (tp->snd_cwnd * (BICTCP_BETA_SCALE + beta))
410 / (2 * BICTCP_BETA_SCALE);
412 ca->last_max_cwnd = tp->snd_cwnd;
414 return max((tp->snd_cwnd * beta) / BICTCP_BETA_SCALE, 2U);
417 void BPF_STRUCT_OPS(bpf_cubic_state, struct sock *sk, __u8 new_state)
419 if (new_state == TCP_CA_Loss) {
420 bictcp_reset(inet_csk_ca(sk));
421 bictcp_hystart_reset(sk);
425 #define GSO_MAX_SIZE 65536
427 /* Account for TSO/GRO delays.
428 * Otherwise short RTT flows could get too small ssthresh, since during
429 * slow start we begin with small TSO packets and ca->delay_min would
430 * not account for long aggregation delay when TSO packets get bigger.
431 * Ideally even with a very small RTT we would like to have at least one
432 * TSO packet being sent and received by GRO, and another one in qdisc layer.
433 * We apply another 100% factor because @rate is doubled at this point.
434 * We cap the cushion to 1ms.
436 static __always_inline __u32 hystart_ack_delay(struct sock *sk)
440 rate = sk->sk_pacing_rate;
443 return min((__u64)USEC_PER_MSEC,
444 div64_ul((__u64)GSO_MAX_SIZE * 4 * USEC_PER_SEC, rate));
447 static __always_inline void hystart_update(struct sock *sk, __u32 delay)
449 struct tcp_sock *tp = tcp_sk(sk);
450 struct bictcp *ca = inet_csk_ca(sk);
453 if (hystart_detect & HYSTART_ACK_TRAIN) {
454 __u32 now = bictcp_clock_us(sk);
456 /* first detection parameter - ack-train detection */
457 if ((__s32)(now - ca->last_ack) <= hystart_ack_delta_us) {
460 threshold = ca->delay_min + hystart_ack_delay(sk);
462 /* Hystart ack train triggers if we get ack past
464 * Pacing might have delayed packets up to RTT/2
467 if (sk->sk_pacing_status == SK_PACING_NONE)
470 if ((__s32)(now - ca->round_start) > threshold) {
472 tp->snd_ssthresh = tp->snd_cwnd;
477 if (hystart_detect & HYSTART_DELAY) {
478 /* obtain the minimum delay of more than sampling packets */
479 if (ca->curr_rtt > delay)
480 ca->curr_rtt = delay;
481 if (ca->sample_cnt < HYSTART_MIN_SAMPLES) {
484 if (ca->curr_rtt > ca->delay_min +
485 HYSTART_DELAY_THRESH(ca->delay_min >> 3)) {
487 tp->snd_ssthresh = tp->snd_cwnd;
493 int bpf_cubic_acked_called = 0;
495 void BPF_STRUCT_OPS(bpf_cubic_acked, struct sock *sk,
496 const struct ack_sample *sample)
498 const struct tcp_sock *tp = tcp_sk(sk);
499 struct bictcp *ca = inet_csk_ca(sk);
502 bpf_cubic_acked_called = 1;
503 /* Some calls are for duplicates without timetamps */
504 if (sample->rtt_us < 0)
507 /* Discard delay samples right after fast recovery */
508 if (ca->epoch_start && (__s32)(tcp_jiffies32 - ca->epoch_start) < HZ)
511 delay = sample->rtt_us;
515 /* first time call or link delay decreases */
516 if (ca->delay_min == 0 || ca->delay_min > delay)
517 ca->delay_min = delay;
519 /* hystart triggers when cwnd is larger than some threshold */
520 if (!ca->found && tcp_in_slow_start(tp) && hystart &&
521 tp->snd_cwnd >= hystart_low_window)
522 hystart_update(sk, delay);
525 extern __u32 tcp_reno_undo_cwnd(struct sock *sk) __ksym;
527 __u32 BPF_STRUCT_OPS(bpf_cubic_undo_cwnd, struct sock *sk)
529 return tcp_reno_undo_cwnd(sk);
533 struct tcp_congestion_ops cubic = {
534 .init = (void *)bpf_cubic_init,
535 .ssthresh = (void *)bpf_cubic_recalc_ssthresh,
536 .cong_avoid = (void *)bpf_cubic_cong_avoid,
537 .set_state = (void *)bpf_cubic_state,
538 .undo_cwnd = (void *)bpf_cubic_undo_cwnd,
539 .cwnd_event = (void *)bpf_cubic_cwnd_event,
540 .pkts_acked = (void *)bpf_cubic_acked,