1 /*- 2 * Copyright (c) 2009-2013 3 * Swinburne University of Technology, Melbourne, Australia 4 * All rights reserved. 5 * 6 * This software was developed at the Centre for Advanced Internet 7 * Architectures, Swinburne University of Technology, by David Hayes, made 8 * possible in part by a gift from The Cisco University Research Program Fund, 9 * a corporate advised fund of Silicon Valley Community Foundation. Development 10 * and testing were further assisted by a grant from the FreeBSD Foundation. 11 * 12 * Redistribution and use in source and binary forms, with or without 13 * modification, are permitted provided that the following conditions 14 * are met: 15 * 1. Redistributions of source code must retain the above copyright 16 * notice, this list of conditions and the following disclaimer. 17 * 2. Redistributions in binary form must reproduce the above copyright 18 * notice, this list of conditions and the following disclaimer in the 19 * documentation and/or other materials provided with the distribution. 20 * 21 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 22 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 23 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 24 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 25 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 26 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 27 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 28 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 29 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 30 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 31 * SUCH DAMAGE. 32 */ 33 34 /* 35 * CAIA Delay-Gradient (CDG) congestion control algorithm 36 * 37 * An implemention of the delay-gradient congestion control algorithm proposed 38 * in the following paper: 39 * 40 * D. A. Hayes and G. Armitage, "Revisiting TCP Congestion Control using Delay 41 * Gradients", in IFIP Networking, Valencia, Spain, 9-13 May 2011. 42 * 43 * Developed as part of the NewTCP research project at Swinburne University of 44 * Technology's Centre for Advanced Internet Architectures, Melbourne, 45 * Australia. More details are available at: 46 * http://caia.swin.edu.au/urp/newtcp/ 47 */ 48 49 #include <sys/cdefs.h> 50 __FBSDID("$FreeBSD$"); 51 52 #include <sys/param.h> 53 #include <sys/hhook.h> 54 #include <sys/kernel.h> 55 #include <sys/khelp.h> 56 #include <sys/limits.h> 57 #include <sys/lock.h> 58 #include <sys/malloc.h> 59 #include <sys/module.h> 60 #include <sys/queue.h> 61 #include <sys/socket.h> 62 #include <sys/socketvar.h> 63 #include <sys/sysctl.h> 64 #include <sys/systm.h> 65 66 #include <net/if.h> 67 #include <net/vnet.h> 68 69 #include <netinet/cc.h> 70 #include <netinet/tcp_seq.h> 71 #include <netinet/tcp_timer.h> 72 #include <netinet/tcp_var.h> 73 74 #include <netinet/cc/cc_module.h> 75 76 #include <netinet/khelp/h_ertt.h> 77 78 #include <vm/uma.h> 79 80 #define CDG_VERSION "0.1" 81 82 #define CAST_PTR_INT(X) (*((int*)(X))) 83 84 #ifndef VIMAGE 85 #define vnet_sysctl_handle_uint(oidp, arg1, arg2, req) \ 86 sysctl_handle_int(oidp, arg1, arg2, req) 87 #endif 88 89 /* Private delay-gradient induced congestion control signal. */ 90 #define CC_CDG_DELAY 0x01000000 91 92 /* NewReno window deflation factor on loss (as a percentage). */ 93 #define RENO_BETA 50 94 95 /* Queue states. */ 96 #define CDG_Q_EMPTY 1 97 #define CDG_Q_RISING 2 98 #define CDG_Q_FALLING 3 99 #define CDG_Q_FULL 4 100 #define CDG_Q_UNKNOWN 9999 101 102 /* Number of bit shifts used in probexp lookup table. */ 103 #define EXP_PREC 15 104 105 /* Largest gradient represented in probexp lookup table. */ 106 #define MAXGRAD 5 107 108 /* 109 * Delay Precision Enhance - number of bit shifts used for qtrend related 110 * integer arithmetic precision. 111 */ 112 #define D_P_E 7 113 114 struct qdiff_sample { 115 long qdiff; 116 STAILQ_ENTRY(qdiff_sample) qdiff_lnk; 117 }; 118 119 struct cdg { 120 long max_qtrend; 121 long min_qtrend; 122 STAILQ_HEAD(minrtts_head, qdiff_sample) qdiffmin_q; 123 STAILQ_HEAD(maxrtts_head, qdiff_sample) qdiffmax_q; 124 long window_incr; 125 /* rttcount for window increase when in congestion avoidance */ 126 long rtt_count; 127 /* maximum measured rtt within an rtt period */ 128 int maxrtt_in_rtt; 129 /* maximum measured rtt within prev rtt period */ 130 int maxrtt_in_prevrtt; 131 /* minimum measured rtt within an rtt period */ 132 int minrtt_in_rtt; 133 /* minimum measured rtt within prev rtt period */ 134 int minrtt_in_prevrtt; 135 /* consecutive congestion episode counter */ 136 uint32_t consec_cong_cnt; 137 /* when tracking a new reno type loss window */ 138 uint32_t shadow_w; 139 /* maximum number of samples in the moving average queue */ 140 int sample_q_size; 141 /* number of samples in the moving average queue */ 142 int num_samples; 143 /* estimate of the queue state of the path */ 144 int queue_state; 145 }; 146 147 /* 148 * Lookup table for: 149 * (1 - exp(-x)) << EXP_PREC, where x = [0,MAXGRAD] in 2^-7 increments 150 * 151 * Note: probexp[0] is set to 10 (not 0) as a safety for very low increase 152 * gradients. 153 */ 154 static const int probexp[641] = { 155 10,255,508,759,1008,1255,1501,1744,1985,2225,2463,2698,2932,3165,3395,3624, 156 3850,4075,4299,4520,4740,4958,5175,5389,5602,5814,6024,6232,6438,6643,6846, 157 7048,7248,7447,7644,7839,8033,8226,8417,8606,8794,8981,9166,9350,9532,9713, 158 9892,10070,10247,10422,10596,10769,10940,11110,11278,11445,11611,11776,11939, 159 12101,12262,12422,12580,12737,12893,13048,13201,13354,13505,13655,13803,13951, 160 14097,14243,14387,14530,14672,14813,14952,15091,15229,15365,15500,15635,15768, 161 15900,16032,16162,16291,16419,16547,16673,16798,16922,17046,17168,17289,17410, 162 17529,17648,17766,17882,17998,18113,18227,18340,18453,18564,18675,18784,18893, 163 19001,19108,19215,19320,19425,19529,19632,19734,19835,19936,20036,20135,20233, 164 20331,20427,20523,20619,20713,20807,20900,20993,21084,21175,21265,21355,21444, 165 21532,21619,21706,21792,21878,21962,22046,22130,22213,22295,22376,22457,22537, 166 22617,22696,22774,22852,22929,23006,23082,23157,23232,23306,23380,23453,23525, 167 23597,23669,23739,23810,23879,23949,24017,24085,24153,24220,24286,24352,24418, 168 24483,24547,24611,24675,24738,24800,24862,24924,24985,25045,25106,25165,25224, 169 25283,25341,25399,25456,25513,25570,25626,25681,25737,25791,25846,25899,25953, 170 26006,26059,26111,26163,26214,26265,26316,26366,26416,26465,26514,26563,26611, 171 26659,26707,26754,26801,26847,26893,26939,26984,27029,27074,27118,27162,27206, 172 27249,27292,27335,27377,27419,27460,27502,27543,27583,27624,27664,27703,27743, 173 27782,27821,27859,27897,27935,27973,28010,28047,28084,28121,28157,28193,28228, 174 28263,28299,28333,28368,28402,28436,28470,28503,28536,28569,28602,28634,28667, 175 28699,28730,28762,28793,28824,28854,28885,28915,28945,28975,29004,29034,29063, 176 29092,29120,29149,29177,29205,29232,29260,29287,29314,29341,29368,29394,29421, 177 29447,29472,29498,29524,29549,29574,29599,29623,29648,29672,29696,29720,29744, 178 29767,29791,29814,29837,29860,29882,29905,29927,29949,29971,29993,30014,30036, 179 30057,30078,30099,30120,30141,30161,30181,30201,30221,30241,30261,30280,30300, 180 30319,30338,30357,30376,30394,30413,30431,30449,30467,30485,30503,30521,30538, 181 30555,30573,30590,30607,30624,30640,30657,30673,30690,30706,30722,30738,30753, 182 30769,30785,30800,30815,30831,30846,30861,30876,30890,30905,30919,30934,30948, 183 30962,30976,30990,31004,31018,31031,31045,31058,31072,31085,31098,31111,31124, 184 31137,31149,31162,31174,31187,31199,31211,31223,31235,31247,31259,31271,31283, 185 31294,31306,31317,31328,31339,31351,31362,31373,31383,31394,31405,31416,31426, 186 31436,31447,31457,31467,31477,31487,31497,31507,31517,31527,31537,31546,31556, 187 31565,31574,31584,31593,31602,31611,31620,31629,31638,31647,31655,31664,31673, 188 31681,31690,31698,31706,31715,31723,31731,31739,31747,31755,31763,31771,31778, 189 31786,31794,31801,31809,31816,31824,31831,31838,31846,31853,31860,31867,31874, 190 31881,31888,31895,31902,31908,31915,31922,31928,31935,31941,31948,31954,31960, 191 31967,31973,31979,31985,31991,31997,32003,32009,32015,32021,32027,32033,32038, 192 32044,32050,32055,32061,32066,32072,32077,32083,32088,32093,32098,32104,32109, 193 32114,32119,32124,32129,32134,32139,32144,32149,32154,32158,32163,32168,32173, 194 32177,32182,32186,32191,32195,32200,32204,32209,32213,32217,32222,32226,32230, 195 32234,32238,32242,32247,32251,32255,32259,32263,32267,32270,32274,32278,32282, 196 32286,32290,32293,32297,32301,32304,32308,32311,32315,32318,32322,32325,32329, 197 32332,32336,32339,32342,32346,32349,32352,32356,32359,32362,32365,32368,32371, 198 32374,32377,32381,32384,32387,32389,32392,32395,32398,32401,32404,32407,32410, 199 32412,32415,32418,32421,32423,32426,32429,32431,32434,32437,32439,32442,32444, 200 32447,32449,32452,32454,32457,32459,32461,32464,32466,32469,32471,32473,32476, 201 32478,32480,32482,32485,32487,32489,32491,32493,32495,32497,32500,32502,32504, 202 32506,32508,32510,32512,32514,32516,32518,32520,32522,32524,32526,32527,32529, 203 32531,32533,32535,32537,32538,32540,32542,32544,32545,32547}; 204 205 static uma_zone_t qdiffsample_zone; 206 207 static MALLOC_DEFINE(M_CDG, "cdg data", 208 "Per connection data required for the CDG congestion control algorithm"); 209 210 static int ertt_id; 211 212 static VNET_DEFINE(uint32_t, cdg_alpha_inc); 213 static VNET_DEFINE(uint32_t, cdg_beta_delay); 214 static VNET_DEFINE(uint32_t, cdg_beta_loss); 215 static VNET_DEFINE(uint32_t, cdg_smoothing_factor); 216 static VNET_DEFINE(uint32_t, cdg_exp_backoff_scale); 217 static VNET_DEFINE(uint32_t, cdg_consec_cong); 218 static VNET_DEFINE(uint32_t, cdg_hold_backoff); 219 #define V_cdg_alpha_inc VNET(cdg_alpha_inc) 220 #define V_cdg_beta_delay VNET(cdg_beta_delay) 221 #define V_cdg_beta_loss VNET(cdg_beta_loss) 222 #define V_cdg_smoothing_factor VNET(cdg_smoothing_factor) 223 #define V_cdg_exp_backoff_scale VNET(cdg_exp_backoff_scale) 224 #define V_cdg_consec_cong VNET(cdg_consec_cong) 225 #define V_cdg_hold_backoff VNET(cdg_hold_backoff) 226 227 /* Function prototypes. */ 228 static int cdg_mod_init(void); 229 static void cdg_conn_init(struct cc_var *ccv); 230 static int cdg_cb_init(struct cc_var *ccv); 231 static void cdg_cb_destroy(struct cc_var *ccv); 232 static void cdg_cong_signal(struct cc_var *ccv, uint32_t signal_type); 233 static void cdg_ack_received(struct cc_var *ccv, uint16_t ack_type); 234 235 struct cc_algo cdg_cc_algo = { 236 .name = "cdg", 237 .mod_init = cdg_mod_init, 238 .ack_received = cdg_ack_received, 239 .cb_destroy = cdg_cb_destroy, 240 .cb_init = cdg_cb_init, 241 .conn_init = cdg_conn_init, 242 .cong_signal = cdg_cong_signal 243 }; 244 245 /* Vnet created and being initialised. */ 246 static void 247 cdg_init_vnet(const void *unused __unused) 248 { 249 250 V_cdg_alpha_inc = 0; 251 V_cdg_beta_delay = 70; 252 V_cdg_beta_loss = 50; 253 V_cdg_smoothing_factor = 8; 254 V_cdg_exp_backoff_scale = 3; 255 V_cdg_consec_cong = 5; 256 V_cdg_hold_backoff = 5; 257 } 258 259 static int 260 cdg_mod_init(void) 261 { 262 VNET_ITERATOR_DECL(v); 263 264 ertt_id = khelp_get_id("ertt"); 265 if (ertt_id <= 0) 266 return (EINVAL); 267 268 qdiffsample_zone = uma_zcreate("cdg_qdiffsample", 269 sizeof(struct qdiff_sample), NULL, NULL, NULL, NULL, 0, 0); 270 271 VNET_LIST_RLOCK(); 272 VNET_FOREACH(v) { 273 CURVNET_SET(v); 274 cdg_init_vnet(NULL); 275 CURVNET_RESTORE(); 276 } 277 VNET_LIST_RUNLOCK(); 278 279 cdg_cc_algo.post_recovery = newreno_cc_algo.post_recovery; 280 cdg_cc_algo.after_idle = newreno_cc_algo.after_idle; 281 282 return (0); 283 } 284 285 static int 286 cdg_cb_init(struct cc_var *ccv) 287 { 288 struct cdg *cdg_data; 289 290 cdg_data = malloc(sizeof(struct cdg), M_CDG, M_NOWAIT); 291 if (cdg_data == NULL) 292 return (ENOMEM); 293 294 cdg_data->shadow_w = 0; 295 cdg_data->max_qtrend = 0; 296 cdg_data->min_qtrend = 0; 297 cdg_data->queue_state = CDG_Q_UNKNOWN; 298 cdg_data->maxrtt_in_rtt = 0; 299 cdg_data->maxrtt_in_prevrtt = 0; 300 cdg_data->minrtt_in_rtt = INT_MAX; 301 cdg_data->minrtt_in_prevrtt = 0; 302 cdg_data->window_incr = 0; 303 cdg_data->rtt_count = 0; 304 cdg_data->consec_cong_cnt = 0; 305 cdg_data->sample_q_size = V_cdg_smoothing_factor; 306 cdg_data->num_samples = 0; 307 STAILQ_INIT(&cdg_data->qdiffmin_q); 308 STAILQ_INIT(&cdg_data->qdiffmax_q); 309 310 ccv->cc_data = cdg_data; 311 312 return (0); 313 } 314 315 static void 316 cdg_conn_init(struct cc_var *ccv) 317 { 318 struct cdg *cdg_data = ccv->cc_data; 319 320 /* 321 * Initialise the shadow_cwnd in case we are competing with loss based 322 * flows from the start 323 */ 324 cdg_data->shadow_w = CCV(ccv, snd_cwnd); 325 } 326 327 static void 328 cdg_cb_destroy(struct cc_var *ccv) 329 { 330 struct cdg *cdg_data; 331 struct qdiff_sample *qds, *qds_n; 332 333 cdg_data = ccv->cc_data; 334 335 qds = STAILQ_FIRST(&cdg_data->qdiffmin_q); 336 while (qds != NULL) { 337 qds_n = STAILQ_NEXT(qds, qdiff_lnk); 338 uma_zfree(qdiffsample_zone,qds); 339 qds = qds_n; 340 } 341 342 qds = STAILQ_FIRST(&cdg_data->qdiffmax_q); 343 while (qds != NULL) { 344 qds_n = STAILQ_NEXT(qds, qdiff_lnk); 345 uma_zfree(qdiffsample_zone,qds); 346 qds = qds_n; 347 } 348 349 free(ccv->cc_data, M_CDG); 350 } 351 352 static int 353 cdg_beta_handler(SYSCTL_HANDLER_ARGS) 354 { 355 356 if (req->newptr != NULL && 357 (CAST_PTR_INT(req->newptr) == 0 || CAST_PTR_INT(req->newptr) > 100)) 358 return (EINVAL); 359 360 return (vnet_sysctl_handle_uint(oidp, arg1, arg2, req)); 361 } 362 363 static int 364 cdg_exp_backoff_scale_handler(SYSCTL_HANDLER_ARGS) 365 { 366 367 if (req->newptr != NULL && CAST_PTR_INT(req->newptr) < 1) 368 return (EINVAL); 369 370 return (vnet_sysctl_handle_uint(oidp, arg1, arg2, req)); 371 } 372 373 static inline unsigned long 374 cdg_window_decrease(struct cc_var *ccv, unsigned long owin, unsigned int beta) 375 { 376 377 return ((ulmin(CCV(ccv, snd_wnd), owin) * beta) / 100); 378 } 379 380 /* 381 * Window increase function 382 * This window increase function is independent of the initial window size 383 * to ensure small window flows are not discriminated against (i.e. fairness). 384 * It increases at 1pkt/rtt like Reno for alpha_inc rtts, and then 2pkts/rtt for 385 * the next alpha_inc rtts, etc. 386 */ 387 static void 388 cdg_window_increase(struct cc_var *ccv, int new_measurement) 389 { 390 struct cdg *cdg_data; 391 int incr, s_w_incr; 392 393 cdg_data = ccv->cc_data; 394 incr = s_w_incr = 0; 395 396 if (CCV(ccv, snd_cwnd) <= CCV(ccv, snd_ssthresh)) { 397 /* Slow start. */ 398 incr = CCV(ccv, t_maxseg); 399 s_w_incr = incr; 400 cdg_data->window_incr = cdg_data->rtt_count = 0; 401 } else { 402 /* Congestion avoidance. */ 403 if (new_measurement) { 404 s_w_incr = CCV(ccv, t_maxseg); 405 if (V_cdg_alpha_inc == 0) { 406 incr = CCV(ccv, t_maxseg); 407 } else { 408 if (++cdg_data->rtt_count >= V_cdg_alpha_inc) { 409 cdg_data->window_incr++; 410 cdg_data->rtt_count = 0; 411 } 412 incr = CCV(ccv, t_maxseg) * 413 cdg_data->window_incr; 414 } 415 } 416 } 417 418 if (cdg_data->shadow_w > 0) 419 cdg_data->shadow_w = ulmin(cdg_data->shadow_w + s_w_incr, 420 TCP_MAXWIN << CCV(ccv, snd_scale)); 421 422 CCV(ccv, snd_cwnd) = ulmin(CCV(ccv, snd_cwnd) + incr, 423 TCP_MAXWIN << CCV(ccv, snd_scale)); 424 } 425 426 static void 427 cdg_cong_signal(struct cc_var *ccv, uint32_t signal_type) 428 { 429 struct cdg *cdg_data = ccv->cc_data; 430 431 switch(signal_type) { 432 case CC_CDG_DELAY: 433 CCV(ccv, snd_ssthresh) = cdg_window_decrease(ccv, 434 CCV(ccv, snd_cwnd), V_cdg_beta_delay); 435 CCV(ccv, snd_cwnd) = CCV(ccv, snd_ssthresh); 436 CCV(ccv, snd_recover) = CCV(ccv, snd_max); 437 cdg_data->window_incr = cdg_data->rtt_count = 0; 438 ENTER_CONGRECOVERY(CCV(ccv, t_flags)); 439 break; 440 case CC_NDUPACK: 441 /* 442 * If already responding to congestion OR we have guessed no 443 * queue in the path is full. 444 */ 445 if (IN_CONGRECOVERY(CCV(ccv, t_flags)) || 446 cdg_data->queue_state < CDG_Q_FULL) { 447 CCV(ccv, snd_ssthresh) = CCV(ccv, snd_cwnd); 448 CCV(ccv, snd_recover) = CCV(ccv, snd_max); 449 } else { 450 /* 451 * Loss is likely to be congestion related. We have 452 * inferred a queue full state, so have shadow window 453 * react to loss as NewReno would. 454 */ 455 if (cdg_data->shadow_w > 0) 456 cdg_data->shadow_w = cdg_window_decrease(ccv, 457 cdg_data->shadow_w, RENO_BETA); 458 459 CCV(ccv, snd_ssthresh) = ulmax(cdg_data->shadow_w, 460 cdg_window_decrease(ccv, CCV(ccv, snd_cwnd), 461 V_cdg_beta_loss)); 462 463 cdg_data->window_incr = cdg_data->rtt_count = 0; 464 } 465 ENTER_RECOVERY(CCV(ccv, t_flags)); 466 break; 467 default: 468 newreno_cc_algo.cong_signal(ccv, signal_type); 469 break; 470 } 471 } 472 473 /* 474 * Using a negative exponential probabilistic backoff so that sources with 475 * varying RTTs which share the same link will, on average, have the same 476 * probability of backoff over time. 477 * 478 * Prob_backoff = 1 - exp(-qtrend / V_cdg_exp_backoff_scale), where 479 * V_cdg_exp_backoff_scale is the average qtrend for the exponential backoff. 480 */ 481 static inline int 482 prob_backoff(long qtrend) 483 { 484 int backoff, idx, p; 485 486 backoff = (qtrend > ((MAXGRAD * V_cdg_exp_backoff_scale) << D_P_E)); 487 488 if (!backoff) { 489 if (V_cdg_exp_backoff_scale > 1) 490 idx = (qtrend + V_cdg_exp_backoff_scale / 2) / 491 V_cdg_exp_backoff_scale; 492 else 493 idx = qtrend; 494 495 /* Backoff probability proportional to rate of queue growth. */ 496 p = (INT_MAX / (1 << EXP_PREC)) * probexp[idx]; 497 backoff = (random() < p); 498 } 499 500 return (backoff); 501 } 502 503 static inline void 504 calc_moving_average(struct cdg *cdg_data, long qdiff_max, long qdiff_min) 505 { 506 struct qdiff_sample *qds; 507 508 ++cdg_data->num_samples; 509 if (cdg_data->num_samples > cdg_data->sample_q_size) { 510 /* Minimum RTT. */ 511 qds = STAILQ_FIRST(&cdg_data->qdiffmin_q); 512 cdg_data->min_qtrend = cdg_data->min_qtrend + 513 (qdiff_min - qds->qdiff) / cdg_data->sample_q_size; 514 STAILQ_REMOVE_HEAD(&cdg_data->qdiffmin_q, qdiff_lnk); 515 qds->qdiff = qdiff_min; 516 STAILQ_INSERT_TAIL(&cdg_data->qdiffmin_q, qds, qdiff_lnk); 517 518 /* Maximum RTT. */ 519 qds = STAILQ_FIRST(&cdg_data->qdiffmax_q); 520 cdg_data->max_qtrend = cdg_data->max_qtrend + 521 (qdiff_max - qds->qdiff) / cdg_data->sample_q_size; 522 STAILQ_REMOVE_HEAD(&cdg_data->qdiffmax_q, qdiff_lnk); 523 qds->qdiff = qdiff_max; 524 STAILQ_INSERT_TAIL(&cdg_data->qdiffmax_q, qds, qdiff_lnk); 525 --cdg_data->num_samples; 526 } else { 527 qds = uma_zalloc(qdiffsample_zone, M_NOWAIT); 528 if (qds != NULL) { 529 cdg_data->min_qtrend = cdg_data->min_qtrend + 530 qdiff_min / cdg_data->sample_q_size; 531 qds->qdiff = qdiff_min; 532 STAILQ_INSERT_TAIL(&cdg_data->qdiffmin_q, qds, 533 qdiff_lnk); 534 } 535 536 qds = uma_zalloc(qdiffsample_zone, M_NOWAIT); 537 if (qds) { 538 cdg_data->max_qtrend = cdg_data->max_qtrend + 539 qdiff_max / cdg_data->sample_q_size; 540 qds->qdiff = qdiff_max; 541 STAILQ_INSERT_TAIL(&cdg_data->qdiffmax_q, qds, 542 qdiff_lnk); 543 } 544 } 545 } 546 547 static void 548 cdg_ack_received(struct cc_var *ccv, uint16_t ack_type) 549 { 550 struct cdg *cdg_data; 551 struct ertt *e_t; 552 long qdiff_max, qdiff_min; 553 int congestion, new_measurement, slowstart; 554 555 cdg_data = ccv->cc_data; 556 e_t = (struct ertt *)khelp_get_osd(CCV(ccv, osd), ertt_id); 557 new_measurement = e_t->flags & ERTT_NEW_MEASUREMENT; 558 congestion = 0; 559 cdg_data->maxrtt_in_rtt = imax(e_t->rtt, cdg_data->maxrtt_in_rtt); 560 cdg_data->minrtt_in_rtt = imin(e_t->rtt, cdg_data->minrtt_in_rtt); 561 562 if (new_measurement) { 563 slowstart = (CCV(ccv, snd_cwnd) <= CCV(ccv, snd_ssthresh)); 564 /* 565 * Update smoothed gradient measurements. Since we are only 566 * using one measurement per RTT, use max or min rtt_in_rtt. 567 * This is also less noisy than a sample RTT measurement. Max 568 * RTT measurements can have trouble due to OS issues. 569 */ 570 if (cdg_data->maxrtt_in_prevrtt) { 571 qdiff_max = ((long)(cdg_data->maxrtt_in_rtt - 572 cdg_data->maxrtt_in_prevrtt) << D_P_E ); 573 qdiff_min = ((long)(cdg_data->minrtt_in_rtt - 574 cdg_data->minrtt_in_prevrtt) << D_P_E ); 575 576 calc_moving_average(cdg_data, qdiff_max, qdiff_min); 577 578 /* Probabilistic backoff with respect to gradient. */ 579 if (slowstart && qdiff_min > 0) 580 congestion = prob_backoff(qdiff_min); 581 else if (cdg_data->min_qtrend > 0) 582 congestion = prob_backoff(cdg_data->min_qtrend); 583 else if (slowstart && qdiff_max > 0) 584 congestion = prob_backoff(qdiff_max); 585 else if (cdg_data->max_qtrend > 0) 586 congestion = prob_backoff(cdg_data->max_qtrend); 587 588 /* Update estimate of queue state. */ 589 if (cdg_data->min_qtrend > 0 && 590 cdg_data->max_qtrend <= 0) { 591 cdg_data->queue_state = CDG_Q_FULL; 592 } else if (cdg_data->min_qtrend >= 0 && 593 cdg_data->max_qtrend < 0) { 594 cdg_data->queue_state = CDG_Q_EMPTY; 595 cdg_data->shadow_w = 0; 596 } else if (cdg_data->min_qtrend > 0 && 597 cdg_data->max_qtrend > 0) { 598 cdg_data->queue_state = CDG_Q_RISING; 599 } else if (cdg_data->min_qtrend < 0 && 600 cdg_data->max_qtrend < 0) { 601 cdg_data->queue_state = CDG_Q_FALLING; 602 } 603 604 if (cdg_data->min_qtrend < 0 || 605 cdg_data->max_qtrend < 0) 606 cdg_data->consec_cong_cnt = 0; 607 } 608 609 cdg_data->minrtt_in_prevrtt = cdg_data->minrtt_in_rtt; 610 cdg_data->minrtt_in_rtt = INT_MAX; 611 cdg_data->maxrtt_in_prevrtt = cdg_data->maxrtt_in_rtt; 612 cdg_data->maxrtt_in_rtt = 0; 613 e_t->flags &= ~ERTT_NEW_MEASUREMENT; 614 } 615 616 if (congestion) { 617 cdg_data->consec_cong_cnt++; 618 if (!IN_RECOVERY(CCV(ccv, t_flags))) { 619 if (cdg_data->consec_cong_cnt <= V_cdg_consec_cong) 620 cdg_cong_signal(ccv, CC_CDG_DELAY); 621 else 622 /* 623 * We have been backing off but the queue is not 624 * falling. Assume we are competing with 625 * loss-based flows and don't back off for the 626 * next V_cdg_hold_backoff RTT periods. 627 */ 628 if (cdg_data->consec_cong_cnt >= 629 V_cdg_consec_cong + V_cdg_hold_backoff) 630 cdg_data->consec_cong_cnt = 0; 631 632 /* Won't see effect until 2nd RTT. */ 633 cdg_data->maxrtt_in_prevrtt = 0; 634 /* 635 * Resync shadow window in case we are competing with a 636 * loss based flow 637 */ 638 cdg_data->shadow_w = ulmax(CCV(ccv, snd_cwnd), 639 cdg_data->shadow_w); 640 } 641 } else if (ack_type == CC_ACK) 642 cdg_window_increase(ccv, new_measurement); 643 } 644 645 /* When a vnet is created and being initialised, init the per-stack CDG vars. */ 646 VNET_SYSINIT(cdg_init_vnet, SI_SUB_PROTO_BEGIN, SI_ORDER_FIRST, 647 cdg_init_vnet, NULL); 648 649 SYSCTL_DECL(_net_inet_tcp_cc_cdg); 650 SYSCTL_NODE(_net_inet_tcp_cc, OID_AUTO, cdg, CTLFLAG_RW, NULL, 651 "CAIA delay-gradient congestion control related settings"); 652 653 SYSCTL_STRING(_net_inet_tcp_cc_cdg, OID_AUTO, version, 654 CTLFLAG_RD, CDG_VERSION, sizeof(CDG_VERSION) - 1, 655 "Current algorithm/implementation version number"); 656 657 SYSCTL_VNET_UINT(_net_inet_tcp_cc_cdg, OID_AUTO, alpha_inc, 658 CTLFLAG_RW, &VNET_NAME(cdg_alpha_inc), 0, 659 "Increment the window increase factor alpha by 1 MSS segment every " 660 "alpha_inc RTTs during congestion avoidance mode."); 661 662 SYSCTL_VNET_PROC(_net_inet_tcp_cc_cdg, OID_AUTO, beta_delay, 663 CTLTYPE_UINT|CTLFLAG_RW, &VNET_NAME(cdg_beta_delay), 70, 664 &cdg_beta_handler, "IU", 665 "Delay-based window decrease factor as a percentage " 666 "(on delay-based backoff, w = w * beta_delay / 100)"); 667 668 SYSCTL_VNET_PROC(_net_inet_tcp_cc_cdg, OID_AUTO, beta_loss, 669 CTLTYPE_UINT|CTLFLAG_RW, &VNET_NAME(cdg_beta_loss), 50, 670 &cdg_beta_handler, "IU", 671 "Loss-based window decrease factor as a percentage " 672 "(on loss-based backoff, w = w * beta_loss / 100)"); 673 674 SYSCTL_VNET_PROC(_net_inet_tcp_cc_cdg, OID_AUTO, exp_backoff_scale, 675 CTLTYPE_UINT|CTLFLAG_RW, &VNET_NAME(cdg_exp_backoff_scale), 2, 676 &cdg_exp_backoff_scale_handler, "IU", 677 "Scaling parameter for the probabilistic exponential backoff"); 678 679 SYSCTL_VNET_UINT(_net_inet_tcp_cc_cdg, OID_AUTO, smoothing_factor, 680 CTLFLAG_RW, &VNET_NAME(cdg_smoothing_factor), 8, 681 "Number of samples used for moving average smoothing (0 = no smoothing)"); 682 683 SYSCTL_VNET_UINT(_net_inet_tcp_cc_cdg, OID_AUTO, loss_compete_consec_cong, 684 CTLFLAG_RW, &VNET_NAME(cdg_consec_cong), 5, 685 "Number of consecutive delay-gradient based congestion episodes which will " 686 "trigger loss based CC compatibility"); 687 688 SYSCTL_VNET_UINT(_net_inet_tcp_cc_cdg, OID_AUTO, loss_compete_hold_backoff, 689 CTLFLAG_RW, &VNET_NAME(cdg_hold_backoff), 5, 690 "Number of consecutive delay-gradient based congestion episodes to hold " 691 "the window backoff for loss based CC compatibility"); 692 693 DECLARE_CC_MODULE(cdg, &cdg_cc_algo); 694 695 MODULE_DEPEND(cdg, ertt, 1, 1, 1); 696