1 // SPDX-License-Identifier: GPL-2.0-or-later 2 /* 3 * net/sched/sch_htb.c Hierarchical token bucket, feed tree version 4 * 5 * Authors: Martin Devera, <devik@cdi.cz> 6 * 7 * Credits (in time order) for older HTB versions: 8 * Stef Coene <stef.coene@docum.org> 9 * HTB support at LARTC mailing list 10 * Ondrej Kraus, <krauso@barr.cz> 11 * found missing INIT_QDISC(htb) 12 * Vladimir Smelhaus, Aamer Akhter, Bert Hubert 13 * helped a lot to locate nasty class stall bug 14 * Andi Kleen, Jamal Hadi, Bert Hubert 15 * code review and helpful comments on shaping 16 * Tomasz Wrona, <tw@eter.tym.pl> 17 * created test case so that I was able to fix nasty bug 18 * Wilfried Weissmann 19 * spotted bug in dequeue code and helped with fix 20 * Jiri Fojtasek 21 * fixed requeue routine 22 * and many others. thanks. 23 */ 24 #include <linux/module.h> 25 #include <linux/moduleparam.h> 26 #include <linux/types.h> 27 #include <linux/kernel.h> 28 #include <linux/string.h> 29 #include <linux/errno.h> 30 #include <linux/skbuff.h> 31 #include <linux/list.h> 32 #include <linux/compiler.h> 33 #include <linux/rbtree.h> 34 #include <linux/workqueue.h> 35 #include <linux/slab.h> 36 #include <net/netlink.h> 37 #include <net/sch_generic.h> 38 #include <net/pkt_sched.h> 39 #include <net/pkt_cls.h> 40 41 /* HTB algorithm. 42 Author: devik@cdi.cz 43 ======================================================================== 44 HTB is like TBF with multiple classes. It is also similar to CBQ because 45 it allows to assign priority to each class in hierarchy. 46 In fact it is another implementation of Floyd's formal sharing. 47 48 Levels: 49 Each class is assigned level. Leaf has ALWAYS level 0 and root 50 classes have level TC_HTB_MAXDEPTH-1. Interior nodes has level 51 one less than their parent. 52 */ 53 54 static int htb_hysteresis __read_mostly = 0; /* whether to use mode hysteresis for speedup */ 55 #define HTB_VER 0x30011 /* major must be matched with number suplied by TC as version */ 56 57 #if HTB_VER >> 16 != TC_HTB_PROTOVER 58 #error "Mismatched sch_htb.c and pkt_sch.h" 59 #endif 60 61 /* Module parameter and sysfs export */ 62 module_param (htb_hysteresis, int, 0640); 63 MODULE_PARM_DESC(htb_hysteresis, "Hysteresis mode, less CPU load, less accurate"); 64 65 static int htb_rate_est = 0; /* htb classes have a default rate estimator */ 66 module_param(htb_rate_est, int, 0640); 67 MODULE_PARM_DESC(htb_rate_est, "setup a default rate estimator (4sec 16sec) for htb classes"); 68 69 /* used internaly to keep status of single class */ 70 enum htb_cmode { 71 HTB_CANT_SEND, /* class can't send and can't borrow */ 72 HTB_MAY_BORROW, /* class can't send but may borrow */ 73 HTB_CAN_SEND /* class can send */ 74 }; 75 76 struct htb_prio { 77 union { 78 struct rb_root row; 79 struct rb_root feed; 80 }; 81 struct rb_node *ptr; 82 /* When class changes from state 1->2 and disconnects from 83 * parent's feed then we lost ptr value and start from the 84 * first child again. Here we store classid of the 85 * last valid ptr (used when ptr is NULL). 86 */ 87 u32 last_ptr_id; 88 }; 89 90 /* interior & leaf nodes; props specific to leaves are marked L: 91 * To reduce false sharing, place mostly read fields at beginning, 92 * and mostly written ones at the end. 93 */ 94 struct htb_class { 95 struct Qdisc_class_common common; 96 struct psched_ratecfg rate; 97 struct psched_ratecfg ceil; 98 s64 buffer, cbuffer;/* token bucket depth/rate */ 99 s64 mbuffer; /* max wait time */ 100 u32 prio; /* these two are used only by leaves... */ 101 int quantum; /* but stored for parent-to-leaf return */ 102 103 struct tcf_proto __rcu *filter_list; /* class attached filters */ 104 struct tcf_block *block; 105 int filter_cnt; 106 107 int level; /* our level (see above) */ 108 unsigned int children; 109 struct htb_class *parent; /* parent class */ 110 111 struct net_rate_estimator __rcu *rate_est; 112 113 /* 114 * Written often fields 115 */ 116 struct gnet_stats_basic_packed bstats; 117 struct gnet_stats_basic_packed bstats_bias; 118 struct tc_htb_xstats xstats; /* our special stats */ 119 120 /* token bucket parameters */ 121 s64 tokens, ctokens;/* current number of tokens */ 122 s64 t_c; /* checkpoint time */ 123 124 union { 125 struct htb_class_leaf { 126 int deficit[TC_HTB_MAXDEPTH]; 127 struct Qdisc *q; 128 } leaf; 129 struct htb_class_inner { 130 struct htb_prio clprio[TC_HTB_NUMPRIO]; 131 } inner; 132 }; 133 s64 pq_key; 134 135 int prio_activity; /* for which prios are we active */ 136 enum htb_cmode cmode; /* current mode of the class */ 137 struct rb_node pq_node; /* node for event queue */ 138 struct rb_node node[TC_HTB_NUMPRIO]; /* node for self or feed tree */ 139 140 unsigned int drops ____cacheline_aligned_in_smp; 141 unsigned int overlimits; 142 }; 143 144 struct htb_level { 145 struct rb_root wait_pq; 146 struct htb_prio hprio[TC_HTB_NUMPRIO]; 147 }; 148 149 struct htb_sched { 150 struct Qdisc_class_hash clhash; 151 int defcls; /* class where unclassified flows go to */ 152 int rate2quantum; /* quant = rate / rate2quantum */ 153 154 /* filters for qdisc itself */ 155 struct tcf_proto __rcu *filter_list; 156 struct tcf_block *block; 157 158 #define HTB_WARN_TOOMANYEVENTS 0x1 159 unsigned int warned; /* only one warning */ 160 int direct_qlen; 161 struct work_struct work; 162 163 /* non shaped skbs; let them go directly thru */ 164 struct qdisc_skb_head direct_queue; 165 u32 direct_pkts; 166 u32 overlimits; 167 168 struct qdisc_watchdog watchdog; 169 170 s64 now; /* cached dequeue time */ 171 172 /* time of nearest event per level (row) */ 173 s64 near_ev_cache[TC_HTB_MAXDEPTH]; 174 175 int row_mask[TC_HTB_MAXDEPTH]; 176 177 struct htb_level hlevel[TC_HTB_MAXDEPTH]; 178 179 struct Qdisc **direct_qdiscs; 180 unsigned int num_direct_qdiscs; 181 182 bool offload; 183 }; 184 185 /* find class in global hash table using given handle */ 186 static inline struct htb_class *htb_find(u32 handle, struct Qdisc *sch) 187 { 188 struct htb_sched *q = qdisc_priv(sch); 189 struct Qdisc_class_common *clc; 190 191 clc = qdisc_class_find(&q->clhash, handle); 192 if (clc == NULL) 193 return NULL; 194 return container_of(clc, struct htb_class, common); 195 } 196 197 static unsigned long htb_search(struct Qdisc *sch, u32 handle) 198 { 199 return (unsigned long)htb_find(handle, sch); 200 } 201 /** 202 * htb_classify - classify a packet into class 203 * 204 * It returns NULL if the packet should be dropped or -1 if the packet 205 * should be passed directly thru. In all other cases leaf class is returned. 206 * We allow direct class selection by classid in priority. The we examine 207 * filters in qdisc and in inner nodes (if higher filter points to the inner 208 * node). If we end up with classid MAJOR:0 we enqueue the skb into special 209 * internal fifo (direct). These packets then go directly thru. If we still 210 * have no valid leaf we try to use MAJOR:default leaf. It still unsuccessful 211 * then finish and return direct queue. 212 */ 213 #define HTB_DIRECT ((struct htb_class *)-1L) 214 215 static struct htb_class *htb_classify(struct sk_buff *skb, struct Qdisc *sch, 216 int *qerr) 217 { 218 struct htb_sched *q = qdisc_priv(sch); 219 struct htb_class *cl; 220 struct tcf_result res; 221 struct tcf_proto *tcf; 222 int result; 223 224 /* allow to select class by setting skb->priority to valid classid; 225 * note that nfmark can be used too by attaching filter fw with no 226 * rules in it 227 */ 228 if (skb->priority == sch->handle) 229 return HTB_DIRECT; /* X:0 (direct flow) selected */ 230 cl = htb_find(skb->priority, sch); 231 if (cl) { 232 if (cl->level == 0) 233 return cl; 234 /* Start with inner filter chain if a non-leaf class is selected */ 235 tcf = rcu_dereference_bh(cl->filter_list); 236 } else { 237 tcf = rcu_dereference_bh(q->filter_list); 238 } 239 240 *qerr = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; 241 while (tcf && (result = tcf_classify(skb, tcf, &res, false)) >= 0) { 242 #ifdef CONFIG_NET_CLS_ACT 243 switch (result) { 244 case TC_ACT_QUEUED: 245 case TC_ACT_STOLEN: 246 case TC_ACT_TRAP: 247 *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; 248 fallthrough; 249 case TC_ACT_SHOT: 250 return NULL; 251 } 252 #endif 253 cl = (void *)res.class; 254 if (!cl) { 255 if (res.classid == sch->handle) 256 return HTB_DIRECT; /* X:0 (direct flow) */ 257 cl = htb_find(res.classid, sch); 258 if (!cl) 259 break; /* filter selected invalid classid */ 260 } 261 if (!cl->level) 262 return cl; /* we hit leaf; return it */ 263 264 /* we have got inner class; apply inner filter chain */ 265 tcf = rcu_dereference_bh(cl->filter_list); 266 } 267 /* classification failed; try to use default class */ 268 cl = htb_find(TC_H_MAKE(TC_H_MAJ(sch->handle), q->defcls), sch); 269 if (!cl || cl->level) 270 return HTB_DIRECT; /* bad default .. this is safe bet */ 271 return cl; 272 } 273 274 /** 275 * htb_add_to_id_tree - adds class to the round robin list 276 * 277 * Routine adds class to the list (actually tree) sorted by classid. 278 * Make sure that class is not already on such list for given prio. 279 */ 280 static void htb_add_to_id_tree(struct rb_root *root, 281 struct htb_class *cl, int prio) 282 { 283 struct rb_node **p = &root->rb_node, *parent = NULL; 284 285 while (*p) { 286 struct htb_class *c; 287 parent = *p; 288 c = rb_entry(parent, struct htb_class, node[prio]); 289 290 if (cl->common.classid > c->common.classid) 291 p = &parent->rb_right; 292 else 293 p = &parent->rb_left; 294 } 295 rb_link_node(&cl->node[prio], parent, p); 296 rb_insert_color(&cl->node[prio], root); 297 } 298 299 /** 300 * htb_add_to_wait_tree - adds class to the event queue with delay 301 * 302 * The class is added to priority event queue to indicate that class will 303 * change its mode in cl->pq_key microseconds. Make sure that class is not 304 * already in the queue. 305 */ 306 static void htb_add_to_wait_tree(struct htb_sched *q, 307 struct htb_class *cl, s64 delay) 308 { 309 struct rb_node **p = &q->hlevel[cl->level].wait_pq.rb_node, *parent = NULL; 310 311 cl->pq_key = q->now + delay; 312 if (cl->pq_key == q->now) 313 cl->pq_key++; 314 315 /* update the nearest event cache */ 316 if (q->near_ev_cache[cl->level] > cl->pq_key) 317 q->near_ev_cache[cl->level] = cl->pq_key; 318 319 while (*p) { 320 struct htb_class *c; 321 parent = *p; 322 c = rb_entry(parent, struct htb_class, pq_node); 323 if (cl->pq_key >= c->pq_key) 324 p = &parent->rb_right; 325 else 326 p = &parent->rb_left; 327 } 328 rb_link_node(&cl->pq_node, parent, p); 329 rb_insert_color(&cl->pq_node, &q->hlevel[cl->level].wait_pq); 330 } 331 332 /** 333 * htb_next_rb_node - finds next node in binary tree 334 * 335 * When we are past last key we return NULL. 336 * Average complexity is 2 steps per call. 337 */ 338 static inline void htb_next_rb_node(struct rb_node **n) 339 { 340 *n = rb_next(*n); 341 } 342 343 /** 344 * htb_add_class_to_row - add class to its row 345 * 346 * The class is added to row at priorities marked in mask. 347 * It does nothing if mask == 0. 348 */ 349 static inline void htb_add_class_to_row(struct htb_sched *q, 350 struct htb_class *cl, int mask) 351 { 352 q->row_mask[cl->level] |= mask; 353 while (mask) { 354 int prio = ffz(~mask); 355 mask &= ~(1 << prio); 356 htb_add_to_id_tree(&q->hlevel[cl->level].hprio[prio].row, cl, prio); 357 } 358 } 359 360 /* If this triggers, it is a bug in this code, but it need not be fatal */ 361 static void htb_safe_rb_erase(struct rb_node *rb, struct rb_root *root) 362 { 363 if (RB_EMPTY_NODE(rb)) { 364 WARN_ON(1); 365 } else { 366 rb_erase(rb, root); 367 RB_CLEAR_NODE(rb); 368 } 369 } 370 371 372 /** 373 * htb_remove_class_from_row - removes class from its row 374 * 375 * The class is removed from row at priorities marked in mask. 376 * It does nothing if mask == 0. 377 */ 378 static inline void htb_remove_class_from_row(struct htb_sched *q, 379 struct htb_class *cl, int mask) 380 { 381 int m = 0; 382 struct htb_level *hlevel = &q->hlevel[cl->level]; 383 384 while (mask) { 385 int prio = ffz(~mask); 386 struct htb_prio *hprio = &hlevel->hprio[prio]; 387 388 mask &= ~(1 << prio); 389 if (hprio->ptr == cl->node + prio) 390 htb_next_rb_node(&hprio->ptr); 391 392 htb_safe_rb_erase(cl->node + prio, &hprio->row); 393 if (!hprio->row.rb_node) 394 m |= 1 << prio; 395 } 396 q->row_mask[cl->level] &= ~m; 397 } 398 399 /** 400 * htb_activate_prios - creates active classe's feed chain 401 * 402 * The class is connected to ancestors and/or appropriate rows 403 * for priorities it is participating on. cl->cmode must be new 404 * (activated) mode. It does nothing if cl->prio_activity == 0. 405 */ 406 static void htb_activate_prios(struct htb_sched *q, struct htb_class *cl) 407 { 408 struct htb_class *p = cl->parent; 409 long m, mask = cl->prio_activity; 410 411 while (cl->cmode == HTB_MAY_BORROW && p && mask) { 412 m = mask; 413 while (m) { 414 int prio = ffz(~m); 415 m &= ~(1 << prio); 416 417 if (p->inner.clprio[prio].feed.rb_node) 418 /* parent already has its feed in use so that 419 * reset bit in mask as parent is already ok 420 */ 421 mask &= ~(1 << prio); 422 423 htb_add_to_id_tree(&p->inner.clprio[prio].feed, cl, prio); 424 } 425 p->prio_activity |= mask; 426 cl = p; 427 p = cl->parent; 428 429 } 430 if (cl->cmode == HTB_CAN_SEND && mask) 431 htb_add_class_to_row(q, cl, mask); 432 } 433 434 /** 435 * htb_deactivate_prios - remove class from feed chain 436 * 437 * cl->cmode must represent old mode (before deactivation). It does 438 * nothing if cl->prio_activity == 0. Class is removed from all feed 439 * chains and rows. 440 */ 441 static void htb_deactivate_prios(struct htb_sched *q, struct htb_class *cl) 442 { 443 struct htb_class *p = cl->parent; 444 long m, mask = cl->prio_activity; 445 446 while (cl->cmode == HTB_MAY_BORROW && p && mask) { 447 m = mask; 448 mask = 0; 449 while (m) { 450 int prio = ffz(~m); 451 m &= ~(1 << prio); 452 453 if (p->inner.clprio[prio].ptr == cl->node + prio) { 454 /* we are removing child which is pointed to from 455 * parent feed - forget the pointer but remember 456 * classid 457 */ 458 p->inner.clprio[prio].last_ptr_id = cl->common.classid; 459 p->inner.clprio[prio].ptr = NULL; 460 } 461 462 htb_safe_rb_erase(cl->node + prio, 463 &p->inner.clprio[prio].feed); 464 465 if (!p->inner.clprio[prio].feed.rb_node) 466 mask |= 1 << prio; 467 } 468 469 p->prio_activity &= ~mask; 470 cl = p; 471 p = cl->parent; 472 473 } 474 if (cl->cmode == HTB_CAN_SEND && mask) 475 htb_remove_class_from_row(q, cl, mask); 476 } 477 478 static inline s64 htb_lowater(const struct htb_class *cl) 479 { 480 if (htb_hysteresis) 481 return cl->cmode != HTB_CANT_SEND ? -cl->cbuffer : 0; 482 else 483 return 0; 484 } 485 static inline s64 htb_hiwater(const struct htb_class *cl) 486 { 487 if (htb_hysteresis) 488 return cl->cmode == HTB_CAN_SEND ? -cl->buffer : 0; 489 else 490 return 0; 491 } 492 493 494 /** 495 * htb_class_mode - computes and returns current class mode 496 * 497 * It computes cl's mode at time cl->t_c+diff and returns it. If mode 498 * is not HTB_CAN_SEND then cl->pq_key is updated to time difference 499 * from now to time when cl will change its state. 500 * Also it is worth to note that class mode doesn't change simply 501 * at cl->{c,}tokens == 0 but there can rather be hysteresis of 502 * 0 .. -cl->{c,}buffer range. It is meant to limit number of 503 * mode transitions per time unit. The speed gain is about 1/6. 504 */ 505 static inline enum htb_cmode 506 htb_class_mode(struct htb_class *cl, s64 *diff) 507 { 508 s64 toks; 509 510 if ((toks = (cl->ctokens + *diff)) < htb_lowater(cl)) { 511 *diff = -toks; 512 return HTB_CANT_SEND; 513 } 514 515 if ((toks = (cl->tokens + *diff)) >= htb_hiwater(cl)) 516 return HTB_CAN_SEND; 517 518 *diff = -toks; 519 return HTB_MAY_BORROW; 520 } 521 522 /** 523 * htb_change_class_mode - changes classe's mode 524 * 525 * This should be the only way how to change classe's mode under normal 526 * cirsumstances. Routine will update feed lists linkage, change mode 527 * and add class to the wait event queue if appropriate. New mode should 528 * be different from old one and cl->pq_key has to be valid if changing 529 * to mode other than HTB_CAN_SEND (see htb_add_to_wait_tree). 530 */ 531 static void 532 htb_change_class_mode(struct htb_sched *q, struct htb_class *cl, s64 *diff) 533 { 534 enum htb_cmode new_mode = htb_class_mode(cl, diff); 535 536 if (new_mode == cl->cmode) 537 return; 538 539 if (new_mode == HTB_CANT_SEND) { 540 cl->overlimits++; 541 q->overlimits++; 542 } 543 544 if (cl->prio_activity) { /* not necessary: speed optimization */ 545 if (cl->cmode != HTB_CANT_SEND) 546 htb_deactivate_prios(q, cl); 547 cl->cmode = new_mode; 548 if (new_mode != HTB_CANT_SEND) 549 htb_activate_prios(q, cl); 550 } else 551 cl->cmode = new_mode; 552 } 553 554 /** 555 * htb_activate - inserts leaf cl into appropriate active feeds 556 * 557 * Routine learns (new) priority of leaf and activates feed chain 558 * for the prio. It can be called on already active leaf safely. 559 * It also adds leaf into droplist. 560 */ 561 static inline void htb_activate(struct htb_sched *q, struct htb_class *cl) 562 { 563 WARN_ON(cl->level || !cl->leaf.q || !cl->leaf.q->q.qlen); 564 565 if (!cl->prio_activity) { 566 cl->prio_activity = 1 << cl->prio; 567 htb_activate_prios(q, cl); 568 } 569 } 570 571 /** 572 * htb_deactivate - remove leaf cl from active feeds 573 * 574 * Make sure that leaf is active. In the other words it can't be called 575 * with non-active leaf. It also removes class from the drop list. 576 */ 577 static inline void htb_deactivate(struct htb_sched *q, struct htb_class *cl) 578 { 579 WARN_ON(!cl->prio_activity); 580 581 htb_deactivate_prios(q, cl); 582 cl->prio_activity = 0; 583 } 584 585 static int htb_enqueue(struct sk_buff *skb, struct Qdisc *sch, 586 struct sk_buff **to_free) 587 { 588 int ret; 589 unsigned int len = qdisc_pkt_len(skb); 590 struct htb_sched *q = qdisc_priv(sch); 591 struct htb_class *cl = htb_classify(skb, sch, &ret); 592 593 if (cl == HTB_DIRECT) { 594 /* enqueue to helper queue */ 595 if (q->direct_queue.qlen < q->direct_qlen) { 596 __qdisc_enqueue_tail(skb, &q->direct_queue); 597 q->direct_pkts++; 598 } else { 599 return qdisc_drop(skb, sch, to_free); 600 } 601 #ifdef CONFIG_NET_CLS_ACT 602 } else if (!cl) { 603 if (ret & __NET_XMIT_BYPASS) 604 qdisc_qstats_drop(sch); 605 __qdisc_drop(skb, to_free); 606 return ret; 607 #endif 608 } else if ((ret = qdisc_enqueue(skb, cl->leaf.q, 609 to_free)) != NET_XMIT_SUCCESS) { 610 if (net_xmit_drop_count(ret)) { 611 qdisc_qstats_drop(sch); 612 cl->drops++; 613 } 614 return ret; 615 } else { 616 htb_activate(q, cl); 617 } 618 619 sch->qstats.backlog += len; 620 sch->q.qlen++; 621 return NET_XMIT_SUCCESS; 622 } 623 624 static inline void htb_accnt_tokens(struct htb_class *cl, int bytes, s64 diff) 625 { 626 s64 toks = diff + cl->tokens; 627 628 if (toks > cl->buffer) 629 toks = cl->buffer; 630 toks -= (s64) psched_l2t_ns(&cl->rate, bytes); 631 if (toks <= -cl->mbuffer) 632 toks = 1 - cl->mbuffer; 633 634 cl->tokens = toks; 635 } 636 637 static inline void htb_accnt_ctokens(struct htb_class *cl, int bytes, s64 diff) 638 { 639 s64 toks = diff + cl->ctokens; 640 641 if (toks > cl->cbuffer) 642 toks = cl->cbuffer; 643 toks -= (s64) psched_l2t_ns(&cl->ceil, bytes); 644 if (toks <= -cl->mbuffer) 645 toks = 1 - cl->mbuffer; 646 647 cl->ctokens = toks; 648 } 649 650 /** 651 * htb_charge_class - charges amount "bytes" to leaf and ancestors 652 * 653 * Routine assumes that packet "bytes" long was dequeued from leaf cl 654 * borrowing from "level". It accounts bytes to ceil leaky bucket for 655 * leaf and all ancestors and to rate bucket for ancestors at levels 656 * "level" and higher. It also handles possible change of mode resulting 657 * from the update. Note that mode can also increase here (MAY_BORROW to 658 * CAN_SEND) because we can use more precise clock that event queue here. 659 * In such case we remove class from event queue first. 660 */ 661 static void htb_charge_class(struct htb_sched *q, struct htb_class *cl, 662 int level, struct sk_buff *skb) 663 { 664 int bytes = qdisc_pkt_len(skb); 665 enum htb_cmode old_mode; 666 s64 diff; 667 668 while (cl) { 669 diff = min_t(s64, q->now - cl->t_c, cl->mbuffer); 670 if (cl->level >= level) { 671 if (cl->level == level) 672 cl->xstats.lends++; 673 htb_accnt_tokens(cl, bytes, diff); 674 } else { 675 cl->xstats.borrows++; 676 cl->tokens += diff; /* we moved t_c; update tokens */ 677 } 678 htb_accnt_ctokens(cl, bytes, diff); 679 cl->t_c = q->now; 680 681 old_mode = cl->cmode; 682 diff = 0; 683 htb_change_class_mode(q, cl, &diff); 684 if (old_mode != cl->cmode) { 685 if (old_mode != HTB_CAN_SEND) 686 htb_safe_rb_erase(&cl->pq_node, &q->hlevel[cl->level].wait_pq); 687 if (cl->cmode != HTB_CAN_SEND) 688 htb_add_to_wait_tree(q, cl, diff); 689 } 690 691 /* update basic stats except for leaves which are already updated */ 692 if (cl->level) 693 bstats_update(&cl->bstats, skb); 694 695 cl = cl->parent; 696 } 697 } 698 699 /** 700 * htb_do_events - make mode changes to classes at the level 701 * 702 * Scans event queue for pending events and applies them. Returns time of 703 * next pending event (0 for no event in pq, q->now for too many events). 704 * Note: Applied are events whose have cl->pq_key <= q->now. 705 */ 706 static s64 htb_do_events(struct htb_sched *q, const int level, 707 unsigned long start) 708 { 709 /* don't run for longer than 2 jiffies; 2 is used instead of 710 * 1 to simplify things when jiffy is going to be incremented 711 * too soon 712 */ 713 unsigned long stop_at = start + 2; 714 struct rb_root *wait_pq = &q->hlevel[level].wait_pq; 715 716 while (time_before(jiffies, stop_at)) { 717 struct htb_class *cl; 718 s64 diff; 719 struct rb_node *p = rb_first(wait_pq); 720 721 if (!p) 722 return 0; 723 724 cl = rb_entry(p, struct htb_class, pq_node); 725 if (cl->pq_key > q->now) 726 return cl->pq_key; 727 728 htb_safe_rb_erase(p, wait_pq); 729 diff = min_t(s64, q->now - cl->t_c, cl->mbuffer); 730 htb_change_class_mode(q, cl, &diff); 731 if (cl->cmode != HTB_CAN_SEND) 732 htb_add_to_wait_tree(q, cl, diff); 733 } 734 735 /* too much load - let's continue after a break for scheduling */ 736 if (!(q->warned & HTB_WARN_TOOMANYEVENTS)) { 737 pr_warn("htb: too many events!\n"); 738 q->warned |= HTB_WARN_TOOMANYEVENTS; 739 } 740 741 return q->now; 742 } 743 744 /* Returns class->node+prio from id-tree where classe's id is >= id. NULL 745 * is no such one exists. 746 */ 747 static struct rb_node *htb_id_find_next_upper(int prio, struct rb_node *n, 748 u32 id) 749 { 750 struct rb_node *r = NULL; 751 while (n) { 752 struct htb_class *cl = 753 rb_entry(n, struct htb_class, node[prio]); 754 755 if (id > cl->common.classid) { 756 n = n->rb_right; 757 } else if (id < cl->common.classid) { 758 r = n; 759 n = n->rb_left; 760 } else { 761 return n; 762 } 763 } 764 return r; 765 } 766 767 /** 768 * htb_lookup_leaf - returns next leaf class in DRR order 769 * 770 * Find leaf where current feed pointers points to. 771 */ 772 static struct htb_class *htb_lookup_leaf(struct htb_prio *hprio, const int prio) 773 { 774 int i; 775 struct { 776 struct rb_node *root; 777 struct rb_node **pptr; 778 u32 *pid; 779 } stk[TC_HTB_MAXDEPTH], *sp = stk; 780 781 BUG_ON(!hprio->row.rb_node); 782 sp->root = hprio->row.rb_node; 783 sp->pptr = &hprio->ptr; 784 sp->pid = &hprio->last_ptr_id; 785 786 for (i = 0; i < 65535; i++) { 787 if (!*sp->pptr && *sp->pid) { 788 /* ptr was invalidated but id is valid - try to recover 789 * the original or next ptr 790 */ 791 *sp->pptr = 792 htb_id_find_next_upper(prio, sp->root, *sp->pid); 793 } 794 *sp->pid = 0; /* ptr is valid now so that remove this hint as it 795 * can become out of date quickly 796 */ 797 if (!*sp->pptr) { /* we are at right end; rewind & go up */ 798 *sp->pptr = sp->root; 799 while ((*sp->pptr)->rb_left) 800 *sp->pptr = (*sp->pptr)->rb_left; 801 if (sp > stk) { 802 sp--; 803 if (!*sp->pptr) { 804 WARN_ON(1); 805 return NULL; 806 } 807 htb_next_rb_node(sp->pptr); 808 } 809 } else { 810 struct htb_class *cl; 811 struct htb_prio *clp; 812 813 cl = rb_entry(*sp->pptr, struct htb_class, node[prio]); 814 if (!cl->level) 815 return cl; 816 clp = &cl->inner.clprio[prio]; 817 (++sp)->root = clp->feed.rb_node; 818 sp->pptr = &clp->ptr; 819 sp->pid = &clp->last_ptr_id; 820 } 821 } 822 WARN_ON(1); 823 return NULL; 824 } 825 826 /* dequeues packet at given priority and level; call only if 827 * you are sure that there is active class at prio/level 828 */ 829 static struct sk_buff *htb_dequeue_tree(struct htb_sched *q, const int prio, 830 const int level) 831 { 832 struct sk_buff *skb = NULL; 833 struct htb_class *cl, *start; 834 struct htb_level *hlevel = &q->hlevel[level]; 835 struct htb_prio *hprio = &hlevel->hprio[prio]; 836 837 /* look initial class up in the row */ 838 start = cl = htb_lookup_leaf(hprio, prio); 839 840 do { 841 next: 842 if (unlikely(!cl)) 843 return NULL; 844 845 /* class can be empty - it is unlikely but can be true if leaf 846 * qdisc drops packets in enqueue routine or if someone used 847 * graft operation on the leaf since last dequeue; 848 * simply deactivate and skip such class 849 */ 850 if (unlikely(cl->leaf.q->q.qlen == 0)) { 851 struct htb_class *next; 852 htb_deactivate(q, cl); 853 854 /* row/level might become empty */ 855 if ((q->row_mask[level] & (1 << prio)) == 0) 856 return NULL; 857 858 next = htb_lookup_leaf(hprio, prio); 859 860 if (cl == start) /* fix start if we just deleted it */ 861 start = next; 862 cl = next; 863 goto next; 864 } 865 866 skb = cl->leaf.q->dequeue(cl->leaf.q); 867 if (likely(skb != NULL)) 868 break; 869 870 qdisc_warn_nonwc("htb", cl->leaf.q); 871 htb_next_rb_node(level ? &cl->parent->inner.clprio[prio].ptr: 872 &q->hlevel[0].hprio[prio].ptr); 873 cl = htb_lookup_leaf(hprio, prio); 874 875 } while (cl != start); 876 877 if (likely(skb != NULL)) { 878 bstats_update(&cl->bstats, skb); 879 cl->leaf.deficit[level] -= qdisc_pkt_len(skb); 880 if (cl->leaf.deficit[level] < 0) { 881 cl->leaf.deficit[level] += cl->quantum; 882 htb_next_rb_node(level ? &cl->parent->inner.clprio[prio].ptr : 883 &q->hlevel[0].hprio[prio].ptr); 884 } 885 /* this used to be after charge_class but this constelation 886 * gives us slightly better performance 887 */ 888 if (!cl->leaf.q->q.qlen) 889 htb_deactivate(q, cl); 890 htb_charge_class(q, cl, level, skb); 891 } 892 return skb; 893 } 894 895 static struct sk_buff *htb_dequeue(struct Qdisc *sch) 896 { 897 struct sk_buff *skb; 898 struct htb_sched *q = qdisc_priv(sch); 899 int level; 900 s64 next_event; 901 unsigned long start_at; 902 903 /* try to dequeue direct packets as high prio (!) to minimize cpu work */ 904 skb = __qdisc_dequeue_head(&q->direct_queue); 905 if (skb != NULL) { 906 ok: 907 qdisc_bstats_update(sch, skb); 908 qdisc_qstats_backlog_dec(sch, skb); 909 sch->q.qlen--; 910 return skb; 911 } 912 913 if (!sch->q.qlen) 914 goto fin; 915 q->now = ktime_get_ns(); 916 start_at = jiffies; 917 918 next_event = q->now + 5LLU * NSEC_PER_SEC; 919 920 for (level = 0; level < TC_HTB_MAXDEPTH; level++) { 921 /* common case optimization - skip event handler quickly */ 922 int m; 923 s64 event = q->near_ev_cache[level]; 924 925 if (q->now >= event) { 926 event = htb_do_events(q, level, start_at); 927 if (!event) 928 event = q->now + NSEC_PER_SEC; 929 q->near_ev_cache[level] = event; 930 } 931 932 if (next_event > event) 933 next_event = event; 934 935 m = ~q->row_mask[level]; 936 while (m != (int)(-1)) { 937 int prio = ffz(m); 938 939 m |= 1 << prio; 940 skb = htb_dequeue_tree(q, prio, level); 941 if (likely(skb != NULL)) 942 goto ok; 943 } 944 } 945 if (likely(next_event > q->now)) 946 qdisc_watchdog_schedule_ns(&q->watchdog, next_event); 947 else 948 schedule_work(&q->work); 949 fin: 950 return skb; 951 } 952 953 /* reset all classes */ 954 /* always caled under BH & queue lock */ 955 static void htb_reset(struct Qdisc *sch) 956 { 957 struct htb_sched *q = qdisc_priv(sch); 958 struct htb_class *cl; 959 unsigned int i; 960 961 for (i = 0; i < q->clhash.hashsize; i++) { 962 hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { 963 if (cl->level) 964 memset(&cl->inner, 0, sizeof(cl->inner)); 965 else { 966 if (cl->leaf.q && !q->offload) 967 qdisc_reset(cl->leaf.q); 968 } 969 cl->prio_activity = 0; 970 cl->cmode = HTB_CAN_SEND; 971 } 972 } 973 qdisc_watchdog_cancel(&q->watchdog); 974 __qdisc_reset_queue(&q->direct_queue); 975 sch->q.qlen = 0; 976 sch->qstats.backlog = 0; 977 memset(q->hlevel, 0, sizeof(q->hlevel)); 978 memset(q->row_mask, 0, sizeof(q->row_mask)); 979 } 980 981 static const struct nla_policy htb_policy[TCA_HTB_MAX + 1] = { 982 [TCA_HTB_PARMS] = { .len = sizeof(struct tc_htb_opt) }, 983 [TCA_HTB_INIT] = { .len = sizeof(struct tc_htb_glob) }, 984 [TCA_HTB_CTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, 985 [TCA_HTB_RTAB] = { .type = NLA_BINARY, .len = TC_RTAB_SIZE }, 986 [TCA_HTB_DIRECT_QLEN] = { .type = NLA_U32 }, 987 [TCA_HTB_RATE64] = { .type = NLA_U64 }, 988 [TCA_HTB_CEIL64] = { .type = NLA_U64 }, 989 [TCA_HTB_OFFLOAD] = { .type = NLA_FLAG }, 990 }; 991 992 static void htb_work_func(struct work_struct *work) 993 { 994 struct htb_sched *q = container_of(work, struct htb_sched, work); 995 struct Qdisc *sch = q->watchdog.qdisc; 996 997 rcu_read_lock(); 998 __netif_schedule(qdisc_root(sch)); 999 rcu_read_unlock(); 1000 } 1001 1002 static void htb_set_lockdep_class_child(struct Qdisc *q) 1003 { 1004 static struct lock_class_key child_key; 1005 1006 lockdep_set_class(qdisc_lock(q), &child_key); 1007 } 1008 1009 static int htb_offload(struct net_device *dev, struct tc_htb_qopt_offload *opt) 1010 { 1011 return dev->netdev_ops->ndo_setup_tc(dev, TC_SETUP_QDISC_HTB, opt); 1012 } 1013 1014 static int htb_init(struct Qdisc *sch, struct nlattr *opt, 1015 struct netlink_ext_ack *extack) 1016 { 1017 struct net_device *dev = qdisc_dev(sch); 1018 struct tc_htb_qopt_offload offload_opt; 1019 struct htb_sched *q = qdisc_priv(sch); 1020 struct nlattr *tb[TCA_HTB_MAX + 1]; 1021 struct tc_htb_glob *gopt; 1022 unsigned int ntx; 1023 int err; 1024 1025 qdisc_watchdog_init(&q->watchdog, sch); 1026 INIT_WORK(&q->work, htb_work_func); 1027 1028 if (!opt) 1029 return -EINVAL; 1030 1031 err = tcf_block_get(&q->block, &q->filter_list, sch, extack); 1032 if (err) 1033 return err; 1034 1035 err = nla_parse_nested_deprecated(tb, TCA_HTB_MAX, opt, htb_policy, 1036 NULL); 1037 if (err < 0) 1038 return err; 1039 1040 if (!tb[TCA_HTB_INIT]) 1041 return -EINVAL; 1042 1043 gopt = nla_data(tb[TCA_HTB_INIT]); 1044 if (gopt->version != HTB_VER >> 16) 1045 return -EINVAL; 1046 1047 q->offload = nla_get_flag(tb[TCA_HTB_OFFLOAD]); 1048 1049 if (q->offload) { 1050 if (sch->parent != TC_H_ROOT) 1051 return -EOPNOTSUPP; 1052 1053 if (!tc_can_offload(dev) || !dev->netdev_ops->ndo_setup_tc) 1054 return -EOPNOTSUPP; 1055 1056 q->num_direct_qdiscs = dev->real_num_tx_queues; 1057 q->direct_qdiscs = kcalloc(q->num_direct_qdiscs, 1058 sizeof(*q->direct_qdiscs), 1059 GFP_KERNEL); 1060 if (!q->direct_qdiscs) 1061 return -ENOMEM; 1062 } 1063 1064 err = qdisc_class_hash_init(&q->clhash); 1065 if (err < 0) 1066 goto err_free_direct_qdiscs; 1067 1068 qdisc_skb_head_init(&q->direct_queue); 1069 1070 if (tb[TCA_HTB_DIRECT_QLEN]) 1071 q->direct_qlen = nla_get_u32(tb[TCA_HTB_DIRECT_QLEN]); 1072 else 1073 q->direct_qlen = qdisc_dev(sch)->tx_queue_len; 1074 1075 if ((q->rate2quantum = gopt->rate2quantum) < 1) 1076 q->rate2quantum = 1; 1077 q->defcls = gopt->defcls; 1078 1079 if (!q->offload) 1080 return 0; 1081 1082 for (ntx = 0; ntx < q->num_direct_qdiscs; ntx++) { 1083 struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, ntx); 1084 struct Qdisc *qdisc; 1085 1086 qdisc = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, 1087 TC_H_MAKE(sch->handle, 0), extack); 1088 if (!qdisc) { 1089 err = -ENOMEM; 1090 goto err_free_qdiscs; 1091 } 1092 1093 htb_set_lockdep_class_child(qdisc); 1094 q->direct_qdiscs[ntx] = qdisc; 1095 qdisc->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; 1096 } 1097 1098 sch->flags |= TCQ_F_MQROOT; 1099 1100 offload_opt = (struct tc_htb_qopt_offload) { 1101 .command = TC_HTB_CREATE, 1102 .parent_classid = TC_H_MAJ(sch->handle) >> 16, 1103 .classid = TC_H_MIN(q->defcls), 1104 .extack = extack, 1105 }; 1106 err = htb_offload(dev, &offload_opt); 1107 if (err) 1108 goto err_free_qdiscs; 1109 1110 return 0; 1111 1112 err_free_qdiscs: 1113 /* TC_HTB_CREATE call failed, avoid any further calls to the driver. */ 1114 q->offload = false; 1115 1116 for (ntx = 0; ntx < q->num_direct_qdiscs && q->direct_qdiscs[ntx]; 1117 ntx++) 1118 qdisc_put(q->direct_qdiscs[ntx]); 1119 1120 qdisc_class_hash_destroy(&q->clhash); 1121 /* Prevent use-after-free and double-free when htb_destroy gets called. 1122 */ 1123 q->clhash.hash = NULL; 1124 q->clhash.hashsize = 0; 1125 1126 err_free_direct_qdiscs: 1127 kfree(q->direct_qdiscs); 1128 q->direct_qdiscs = NULL; 1129 return err; 1130 } 1131 1132 static void htb_attach_offload(struct Qdisc *sch) 1133 { 1134 struct net_device *dev = qdisc_dev(sch); 1135 struct htb_sched *q = qdisc_priv(sch); 1136 unsigned int ntx; 1137 1138 for (ntx = 0; ntx < q->num_direct_qdiscs; ntx++) { 1139 struct Qdisc *old, *qdisc = q->direct_qdiscs[ntx]; 1140 1141 old = dev_graft_qdisc(qdisc->dev_queue, qdisc); 1142 qdisc_put(old); 1143 qdisc_hash_add(qdisc, false); 1144 } 1145 for (ntx = q->num_direct_qdiscs; ntx < dev->num_tx_queues; ntx++) { 1146 struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, ntx); 1147 struct Qdisc *old = dev_graft_qdisc(dev_queue, NULL); 1148 1149 qdisc_put(old); 1150 } 1151 1152 kfree(q->direct_qdiscs); 1153 q->direct_qdiscs = NULL; 1154 } 1155 1156 static void htb_attach_software(struct Qdisc *sch) 1157 { 1158 struct net_device *dev = qdisc_dev(sch); 1159 unsigned int ntx; 1160 1161 /* Resemble qdisc_graft behavior. */ 1162 for (ntx = 0; ntx < dev->num_tx_queues; ntx++) { 1163 struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, ntx); 1164 struct Qdisc *old = dev_graft_qdisc(dev_queue, sch); 1165 1166 qdisc_refcount_inc(sch); 1167 1168 qdisc_put(old); 1169 } 1170 } 1171 1172 static void htb_attach(struct Qdisc *sch) 1173 { 1174 struct htb_sched *q = qdisc_priv(sch); 1175 1176 if (q->offload) 1177 htb_attach_offload(sch); 1178 else 1179 htb_attach_software(sch); 1180 } 1181 1182 static int htb_dump(struct Qdisc *sch, struct sk_buff *skb) 1183 { 1184 struct htb_sched *q = qdisc_priv(sch); 1185 struct nlattr *nest; 1186 struct tc_htb_glob gopt; 1187 1188 if (q->offload) 1189 sch->flags |= TCQ_F_OFFLOADED; 1190 else 1191 sch->flags &= ~TCQ_F_OFFLOADED; 1192 1193 sch->qstats.overlimits = q->overlimits; 1194 /* Its safe to not acquire qdisc lock. As we hold RTNL, 1195 * no change can happen on the qdisc parameters. 1196 */ 1197 1198 gopt.direct_pkts = q->direct_pkts; 1199 gopt.version = HTB_VER; 1200 gopt.rate2quantum = q->rate2quantum; 1201 gopt.defcls = q->defcls; 1202 gopt.debug = 0; 1203 1204 nest = nla_nest_start_noflag(skb, TCA_OPTIONS); 1205 if (nest == NULL) 1206 goto nla_put_failure; 1207 if (nla_put(skb, TCA_HTB_INIT, sizeof(gopt), &gopt) || 1208 nla_put_u32(skb, TCA_HTB_DIRECT_QLEN, q->direct_qlen)) 1209 goto nla_put_failure; 1210 if (q->offload && nla_put_flag(skb, TCA_HTB_OFFLOAD)) 1211 goto nla_put_failure; 1212 1213 return nla_nest_end(skb, nest); 1214 1215 nla_put_failure: 1216 nla_nest_cancel(skb, nest); 1217 return -1; 1218 } 1219 1220 static int htb_dump_class(struct Qdisc *sch, unsigned long arg, 1221 struct sk_buff *skb, struct tcmsg *tcm) 1222 { 1223 struct htb_class *cl = (struct htb_class *)arg; 1224 struct htb_sched *q = qdisc_priv(sch); 1225 struct nlattr *nest; 1226 struct tc_htb_opt opt; 1227 1228 /* Its safe to not acquire qdisc lock. As we hold RTNL, 1229 * no change can happen on the class parameters. 1230 */ 1231 tcm->tcm_parent = cl->parent ? cl->parent->common.classid : TC_H_ROOT; 1232 tcm->tcm_handle = cl->common.classid; 1233 if (!cl->level && cl->leaf.q) 1234 tcm->tcm_info = cl->leaf.q->handle; 1235 1236 nest = nla_nest_start_noflag(skb, TCA_OPTIONS); 1237 if (nest == NULL) 1238 goto nla_put_failure; 1239 1240 memset(&opt, 0, sizeof(opt)); 1241 1242 psched_ratecfg_getrate(&opt.rate, &cl->rate); 1243 opt.buffer = PSCHED_NS2TICKS(cl->buffer); 1244 psched_ratecfg_getrate(&opt.ceil, &cl->ceil); 1245 opt.cbuffer = PSCHED_NS2TICKS(cl->cbuffer); 1246 opt.quantum = cl->quantum; 1247 opt.prio = cl->prio; 1248 opt.level = cl->level; 1249 if (nla_put(skb, TCA_HTB_PARMS, sizeof(opt), &opt)) 1250 goto nla_put_failure; 1251 if (q->offload && nla_put_flag(skb, TCA_HTB_OFFLOAD)) 1252 goto nla_put_failure; 1253 if ((cl->rate.rate_bytes_ps >= (1ULL << 32)) && 1254 nla_put_u64_64bit(skb, TCA_HTB_RATE64, cl->rate.rate_bytes_ps, 1255 TCA_HTB_PAD)) 1256 goto nla_put_failure; 1257 if ((cl->ceil.rate_bytes_ps >= (1ULL << 32)) && 1258 nla_put_u64_64bit(skb, TCA_HTB_CEIL64, cl->ceil.rate_bytes_ps, 1259 TCA_HTB_PAD)) 1260 goto nla_put_failure; 1261 1262 return nla_nest_end(skb, nest); 1263 1264 nla_put_failure: 1265 nla_nest_cancel(skb, nest); 1266 return -1; 1267 } 1268 1269 static void htb_offload_aggregate_stats(struct htb_sched *q, 1270 struct htb_class *cl) 1271 { 1272 struct htb_class *c; 1273 unsigned int i; 1274 1275 memset(&cl->bstats, 0, sizeof(cl->bstats)); 1276 1277 for (i = 0; i < q->clhash.hashsize; i++) { 1278 hlist_for_each_entry(c, &q->clhash.hash[i], common.hnode) { 1279 struct htb_class *p = c; 1280 1281 while (p && p->level < cl->level) 1282 p = p->parent; 1283 1284 if (p != cl) 1285 continue; 1286 1287 cl->bstats.bytes += c->bstats_bias.bytes; 1288 cl->bstats.packets += c->bstats_bias.packets; 1289 if (c->level == 0) { 1290 cl->bstats.bytes += c->leaf.q->bstats.bytes; 1291 cl->bstats.packets += c->leaf.q->bstats.packets; 1292 } 1293 } 1294 } 1295 } 1296 1297 static int 1298 htb_dump_class_stats(struct Qdisc *sch, unsigned long arg, struct gnet_dump *d) 1299 { 1300 struct htb_class *cl = (struct htb_class *)arg; 1301 struct htb_sched *q = qdisc_priv(sch); 1302 struct gnet_stats_queue qs = { 1303 .drops = cl->drops, 1304 .overlimits = cl->overlimits, 1305 }; 1306 __u32 qlen = 0; 1307 1308 if (!cl->level && cl->leaf.q) 1309 qdisc_qstats_qlen_backlog(cl->leaf.q, &qlen, &qs.backlog); 1310 1311 cl->xstats.tokens = clamp_t(s64, PSCHED_NS2TICKS(cl->tokens), 1312 INT_MIN, INT_MAX); 1313 cl->xstats.ctokens = clamp_t(s64, PSCHED_NS2TICKS(cl->ctokens), 1314 INT_MIN, INT_MAX); 1315 1316 if (q->offload) { 1317 if (!cl->level) { 1318 if (cl->leaf.q) 1319 cl->bstats = cl->leaf.q->bstats; 1320 else 1321 memset(&cl->bstats, 0, sizeof(cl->bstats)); 1322 cl->bstats.bytes += cl->bstats_bias.bytes; 1323 cl->bstats.packets += cl->bstats_bias.packets; 1324 } else { 1325 htb_offload_aggregate_stats(q, cl); 1326 } 1327 } 1328 1329 if (gnet_stats_copy_basic(qdisc_root_sleeping_running(sch), 1330 d, NULL, &cl->bstats) < 0 || 1331 gnet_stats_copy_rate_est(d, &cl->rate_est) < 0 || 1332 gnet_stats_copy_queue(d, NULL, &qs, qlen) < 0) 1333 return -1; 1334 1335 return gnet_stats_copy_app(d, &cl->xstats, sizeof(cl->xstats)); 1336 } 1337 1338 static struct netdev_queue * 1339 htb_select_queue(struct Qdisc *sch, struct tcmsg *tcm) 1340 { 1341 struct net_device *dev = qdisc_dev(sch); 1342 struct tc_htb_qopt_offload offload_opt; 1343 int err; 1344 1345 offload_opt = (struct tc_htb_qopt_offload) { 1346 .command = TC_HTB_LEAF_QUERY_QUEUE, 1347 .classid = TC_H_MIN(tcm->tcm_parent), 1348 }; 1349 err = htb_offload(dev, &offload_opt); 1350 if (err || offload_opt.qid >= dev->num_tx_queues) 1351 return NULL; 1352 return netdev_get_tx_queue(dev, offload_opt.qid); 1353 } 1354 1355 static struct Qdisc * 1356 htb_graft_helper(struct netdev_queue *dev_queue, struct Qdisc *new_q) 1357 { 1358 struct net_device *dev = dev_queue->dev; 1359 struct Qdisc *old_q; 1360 1361 if (dev->flags & IFF_UP) 1362 dev_deactivate(dev); 1363 old_q = dev_graft_qdisc(dev_queue, new_q); 1364 if (new_q) 1365 new_q->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT; 1366 if (dev->flags & IFF_UP) 1367 dev_activate(dev); 1368 1369 return old_q; 1370 } 1371 1372 static void htb_offload_move_qdisc(struct Qdisc *sch, u16 qid_old, u16 qid_new) 1373 { 1374 struct netdev_queue *queue_old, *queue_new; 1375 struct net_device *dev = qdisc_dev(sch); 1376 struct Qdisc *qdisc; 1377 1378 queue_old = netdev_get_tx_queue(dev, qid_old); 1379 queue_new = netdev_get_tx_queue(dev, qid_new); 1380 1381 if (dev->flags & IFF_UP) 1382 dev_deactivate(dev); 1383 qdisc = dev_graft_qdisc(queue_old, NULL); 1384 qdisc->dev_queue = queue_new; 1385 qdisc = dev_graft_qdisc(queue_new, qdisc); 1386 if (dev->flags & IFF_UP) 1387 dev_activate(dev); 1388 1389 WARN_ON(!(qdisc->flags & TCQ_F_BUILTIN)); 1390 } 1391 1392 static int htb_graft(struct Qdisc *sch, unsigned long arg, struct Qdisc *new, 1393 struct Qdisc **old, struct netlink_ext_ack *extack) 1394 { 1395 struct netdev_queue *dev_queue = sch->dev_queue; 1396 struct htb_class *cl = (struct htb_class *)arg; 1397 struct htb_sched *q = qdisc_priv(sch); 1398 struct Qdisc *old_q; 1399 1400 if (cl->level) 1401 return -EINVAL; 1402 1403 if (q->offload) { 1404 dev_queue = new->dev_queue; 1405 WARN_ON(dev_queue != cl->leaf.q->dev_queue); 1406 } 1407 1408 if (!new) { 1409 new = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, 1410 cl->common.classid, extack); 1411 if (!new) 1412 return -ENOBUFS; 1413 } 1414 1415 if (q->offload) { 1416 htb_set_lockdep_class_child(new); 1417 /* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ 1418 qdisc_refcount_inc(new); 1419 old_q = htb_graft_helper(dev_queue, new); 1420 } 1421 1422 *old = qdisc_replace(sch, new, &cl->leaf.q); 1423 1424 if (q->offload) { 1425 WARN_ON(old_q != *old); 1426 qdisc_put(old_q); 1427 } 1428 1429 return 0; 1430 } 1431 1432 static struct Qdisc *htb_leaf(struct Qdisc *sch, unsigned long arg) 1433 { 1434 struct htb_class *cl = (struct htb_class *)arg; 1435 return !cl->level ? cl->leaf.q : NULL; 1436 } 1437 1438 static void htb_qlen_notify(struct Qdisc *sch, unsigned long arg) 1439 { 1440 struct htb_class *cl = (struct htb_class *)arg; 1441 1442 htb_deactivate(qdisc_priv(sch), cl); 1443 } 1444 1445 static inline int htb_parent_last_child(struct htb_class *cl) 1446 { 1447 if (!cl->parent) 1448 /* the root class */ 1449 return 0; 1450 if (cl->parent->children > 1) 1451 /* not the last child */ 1452 return 0; 1453 return 1; 1454 } 1455 1456 static void htb_parent_to_leaf(struct Qdisc *sch, struct htb_class *cl, 1457 struct Qdisc *new_q) 1458 { 1459 struct htb_sched *q = qdisc_priv(sch); 1460 struct htb_class *parent = cl->parent; 1461 1462 WARN_ON(cl->level || !cl->leaf.q || cl->prio_activity); 1463 1464 if (parent->cmode != HTB_CAN_SEND) 1465 htb_safe_rb_erase(&parent->pq_node, 1466 &q->hlevel[parent->level].wait_pq); 1467 1468 parent->level = 0; 1469 memset(&parent->inner, 0, sizeof(parent->inner)); 1470 parent->leaf.q = new_q ? new_q : &noop_qdisc; 1471 parent->tokens = parent->buffer; 1472 parent->ctokens = parent->cbuffer; 1473 parent->t_c = ktime_get_ns(); 1474 parent->cmode = HTB_CAN_SEND; 1475 } 1476 1477 static void htb_parent_to_leaf_offload(struct Qdisc *sch, 1478 struct netdev_queue *dev_queue, 1479 struct Qdisc *new_q) 1480 { 1481 struct Qdisc *old_q; 1482 1483 /* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ 1484 qdisc_refcount_inc(new_q); 1485 old_q = htb_graft_helper(dev_queue, new_q); 1486 WARN_ON(!(old_q->flags & TCQ_F_BUILTIN)); 1487 } 1488 1489 static int htb_destroy_class_offload(struct Qdisc *sch, struct htb_class *cl, 1490 bool last_child, bool destroying, 1491 struct netlink_ext_ack *extack) 1492 { 1493 struct tc_htb_qopt_offload offload_opt; 1494 struct Qdisc *q = cl->leaf.q; 1495 struct Qdisc *old = NULL; 1496 int err; 1497 1498 if (cl->level) 1499 return -EINVAL; 1500 1501 WARN_ON(!q); 1502 if (!destroying) { 1503 /* On destroy of HTB, two cases are possible: 1504 * 1. q is a normal qdisc, but q->dev_queue has noop qdisc. 1505 * 2. q is a noop qdisc (for nodes that were inner), 1506 * q->dev_queue is noop_netdev_queue. 1507 */ 1508 old = htb_graft_helper(q->dev_queue, NULL); 1509 WARN_ON(!old); 1510 WARN_ON(old != q); 1511 } 1512 1513 if (cl->parent) { 1514 cl->parent->bstats_bias.bytes += q->bstats.bytes; 1515 cl->parent->bstats_bias.packets += q->bstats.packets; 1516 } 1517 1518 offload_opt = (struct tc_htb_qopt_offload) { 1519 .command = !last_child ? TC_HTB_LEAF_DEL : 1520 destroying ? TC_HTB_LEAF_DEL_LAST_FORCE : 1521 TC_HTB_LEAF_DEL_LAST, 1522 .classid = cl->common.classid, 1523 .extack = extack, 1524 }; 1525 err = htb_offload(qdisc_dev(sch), &offload_opt); 1526 1527 if (!err || destroying) 1528 qdisc_put(old); 1529 else 1530 htb_graft_helper(q->dev_queue, old); 1531 1532 if (last_child) 1533 return err; 1534 1535 if (!err && offload_opt.moved_qid != 0) { 1536 if (destroying) 1537 q->dev_queue = netdev_get_tx_queue(qdisc_dev(sch), 1538 offload_opt.qid); 1539 else 1540 htb_offload_move_qdisc(sch, offload_opt.moved_qid, 1541 offload_opt.qid); 1542 } 1543 1544 return err; 1545 } 1546 1547 static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl) 1548 { 1549 if (!cl->level) { 1550 WARN_ON(!cl->leaf.q); 1551 qdisc_put(cl->leaf.q); 1552 } 1553 gen_kill_estimator(&cl->rate_est); 1554 tcf_block_put(cl->block); 1555 kfree(cl); 1556 } 1557 1558 static void htb_destroy(struct Qdisc *sch) 1559 { 1560 struct net_device *dev = qdisc_dev(sch); 1561 struct tc_htb_qopt_offload offload_opt; 1562 struct htb_sched *q = qdisc_priv(sch); 1563 struct hlist_node *next; 1564 bool nonempty, changed; 1565 struct htb_class *cl; 1566 unsigned int i; 1567 1568 cancel_work_sync(&q->work); 1569 qdisc_watchdog_cancel(&q->watchdog); 1570 /* This line used to be after htb_destroy_class call below 1571 * and surprisingly it worked in 2.4. But it must precede it 1572 * because filter need its target class alive to be able to call 1573 * unbind_filter on it (without Oops). 1574 */ 1575 tcf_block_put(q->block); 1576 1577 for (i = 0; i < q->clhash.hashsize; i++) { 1578 hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { 1579 tcf_block_put(cl->block); 1580 cl->block = NULL; 1581 } 1582 } 1583 1584 do { 1585 nonempty = false; 1586 changed = false; 1587 for (i = 0; i < q->clhash.hashsize; i++) { 1588 hlist_for_each_entry_safe(cl, next, &q->clhash.hash[i], 1589 common.hnode) { 1590 bool last_child; 1591 1592 if (!q->offload) { 1593 htb_destroy_class(sch, cl); 1594 continue; 1595 } 1596 1597 nonempty = true; 1598 1599 if (cl->level) 1600 continue; 1601 1602 changed = true; 1603 1604 last_child = htb_parent_last_child(cl); 1605 htb_destroy_class_offload(sch, cl, last_child, 1606 true, NULL); 1607 qdisc_class_hash_remove(&q->clhash, 1608 &cl->common); 1609 if (cl->parent) 1610 cl->parent->children--; 1611 if (last_child) 1612 htb_parent_to_leaf(sch, cl, NULL); 1613 htb_destroy_class(sch, cl); 1614 } 1615 } 1616 } while (changed); 1617 WARN_ON(nonempty); 1618 1619 qdisc_class_hash_destroy(&q->clhash); 1620 __qdisc_reset_queue(&q->direct_queue); 1621 1622 if (!q->offload) 1623 return; 1624 1625 offload_opt = (struct tc_htb_qopt_offload) { 1626 .command = TC_HTB_DESTROY, 1627 }; 1628 htb_offload(dev, &offload_opt); 1629 1630 if (!q->direct_qdiscs) 1631 return; 1632 for (i = 0; i < q->num_direct_qdiscs && q->direct_qdiscs[i]; i++) 1633 qdisc_put(q->direct_qdiscs[i]); 1634 kfree(q->direct_qdiscs); 1635 } 1636 1637 static int htb_delete(struct Qdisc *sch, unsigned long arg, 1638 struct netlink_ext_ack *extack) 1639 { 1640 struct htb_sched *q = qdisc_priv(sch); 1641 struct htb_class *cl = (struct htb_class *)arg; 1642 struct Qdisc *new_q = NULL; 1643 int last_child = 0; 1644 int err; 1645 1646 /* TODO: why don't allow to delete subtree ? references ? does 1647 * tc subsys guarantee us that in htb_destroy it holds no class 1648 * refs so that we can remove children safely there ? 1649 */ 1650 if (cl->children || cl->filter_cnt) 1651 return -EBUSY; 1652 1653 if (!cl->level && htb_parent_last_child(cl)) 1654 last_child = 1; 1655 1656 if (q->offload) { 1657 err = htb_destroy_class_offload(sch, cl, last_child, false, 1658 extack); 1659 if (err) 1660 return err; 1661 } 1662 1663 if (last_child) { 1664 struct netdev_queue *dev_queue; 1665 1666 dev_queue = q->offload ? cl->leaf.q->dev_queue : sch->dev_queue; 1667 new_q = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, 1668 cl->parent->common.classid, 1669 NULL); 1670 if (q->offload) { 1671 if (new_q) 1672 htb_set_lockdep_class_child(new_q); 1673 htb_parent_to_leaf_offload(sch, dev_queue, new_q); 1674 } 1675 } 1676 1677 sch_tree_lock(sch); 1678 1679 if (!cl->level) 1680 qdisc_purge_queue(cl->leaf.q); 1681 1682 /* delete from hash and active; remainder in destroy_class */ 1683 qdisc_class_hash_remove(&q->clhash, &cl->common); 1684 if (cl->parent) 1685 cl->parent->children--; 1686 1687 if (cl->prio_activity) 1688 htb_deactivate(q, cl); 1689 1690 if (cl->cmode != HTB_CAN_SEND) 1691 htb_safe_rb_erase(&cl->pq_node, 1692 &q->hlevel[cl->level].wait_pq); 1693 1694 if (last_child) 1695 htb_parent_to_leaf(sch, cl, new_q); 1696 1697 sch_tree_unlock(sch); 1698 1699 htb_destroy_class(sch, cl); 1700 return 0; 1701 } 1702 1703 static int htb_change_class(struct Qdisc *sch, u32 classid, 1704 u32 parentid, struct nlattr **tca, 1705 unsigned long *arg, struct netlink_ext_ack *extack) 1706 { 1707 int err = -EINVAL; 1708 struct htb_sched *q = qdisc_priv(sch); 1709 struct htb_class *cl = (struct htb_class *)*arg, *parent; 1710 struct tc_htb_qopt_offload offload_opt; 1711 struct nlattr *opt = tca[TCA_OPTIONS]; 1712 struct nlattr *tb[TCA_HTB_MAX + 1]; 1713 struct Qdisc *parent_qdisc = NULL; 1714 struct netdev_queue *dev_queue; 1715 struct tc_htb_opt *hopt; 1716 u64 rate64, ceil64; 1717 int warn = 0; 1718 1719 /* extract all subattrs from opt attr */ 1720 if (!opt) 1721 goto failure; 1722 1723 err = nla_parse_nested_deprecated(tb, TCA_HTB_MAX, opt, htb_policy, 1724 NULL); 1725 if (err < 0) 1726 goto failure; 1727 1728 err = -EINVAL; 1729 if (tb[TCA_HTB_PARMS] == NULL) 1730 goto failure; 1731 1732 parent = parentid == TC_H_ROOT ? NULL : htb_find(parentid, sch); 1733 1734 hopt = nla_data(tb[TCA_HTB_PARMS]); 1735 if (!hopt->rate.rate || !hopt->ceil.rate) 1736 goto failure; 1737 1738 /* Keeping backward compatible with rate_table based iproute2 tc */ 1739 if (hopt->rate.linklayer == TC_LINKLAYER_UNAWARE) 1740 qdisc_put_rtab(qdisc_get_rtab(&hopt->rate, tb[TCA_HTB_RTAB], 1741 NULL)); 1742 1743 if (hopt->ceil.linklayer == TC_LINKLAYER_UNAWARE) 1744 qdisc_put_rtab(qdisc_get_rtab(&hopt->ceil, tb[TCA_HTB_CTAB], 1745 NULL)); 1746 1747 rate64 = tb[TCA_HTB_RATE64] ? nla_get_u64(tb[TCA_HTB_RATE64]) : 0; 1748 ceil64 = tb[TCA_HTB_CEIL64] ? nla_get_u64(tb[TCA_HTB_CEIL64]) : 0; 1749 1750 if (!cl) { /* new class */ 1751 struct net_device *dev = qdisc_dev(sch); 1752 struct Qdisc *new_q, *old_q; 1753 int prio; 1754 struct { 1755 struct nlattr nla; 1756 struct gnet_estimator opt; 1757 } est = { 1758 .nla = { 1759 .nla_len = nla_attr_size(sizeof(est.opt)), 1760 .nla_type = TCA_RATE, 1761 }, 1762 .opt = { 1763 /* 4s interval, 16s averaging constant */ 1764 .interval = 2, 1765 .ewma_log = 2, 1766 }, 1767 }; 1768 1769 /* check for valid classid */ 1770 if (!classid || TC_H_MAJ(classid ^ sch->handle) || 1771 htb_find(classid, sch)) 1772 goto failure; 1773 1774 /* check maximal depth */ 1775 if (parent && parent->parent && parent->parent->level < 2) { 1776 pr_err("htb: tree is too deep\n"); 1777 goto failure; 1778 } 1779 err = -ENOBUFS; 1780 cl = kzalloc(sizeof(*cl), GFP_KERNEL); 1781 if (!cl) 1782 goto failure; 1783 1784 err = tcf_block_get(&cl->block, &cl->filter_list, sch, extack); 1785 if (err) { 1786 kfree(cl); 1787 goto failure; 1788 } 1789 if (htb_rate_est || tca[TCA_RATE]) { 1790 err = gen_new_estimator(&cl->bstats, NULL, 1791 &cl->rate_est, 1792 NULL, 1793 qdisc_root_sleeping_running(sch), 1794 tca[TCA_RATE] ? : &est.nla); 1795 if (err) 1796 goto err_block_put; 1797 } 1798 1799 cl->children = 0; 1800 RB_CLEAR_NODE(&cl->pq_node); 1801 1802 for (prio = 0; prio < TC_HTB_NUMPRIO; prio++) 1803 RB_CLEAR_NODE(&cl->node[prio]); 1804 1805 cl->common.classid = classid; 1806 1807 /* Make sure nothing interrupts us in between of two 1808 * ndo_setup_tc calls. 1809 */ 1810 ASSERT_RTNL(); 1811 1812 /* create leaf qdisc early because it uses kmalloc(GFP_KERNEL) 1813 * so that can't be used inside of sch_tree_lock 1814 * -- thanks to Karlis Peisenieks 1815 */ 1816 if (!q->offload) { 1817 dev_queue = sch->dev_queue; 1818 } else if (!(parent && !parent->level)) { 1819 /* Assign a dev_queue to this classid. */ 1820 offload_opt = (struct tc_htb_qopt_offload) { 1821 .command = TC_HTB_LEAF_ALLOC_QUEUE, 1822 .classid = cl->common.classid, 1823 .parent_classid = parent ? 1824 TC_H_MIN(parent->common.classid) : 1825 TC_HTB_CLASSID_ROOT, 1826 .rate = max_t(u64, hopt->rate.rate, rate64), 1827 .ceil = max_t(u64, hopt->ceil.rate, ceil64), 1828 .extack = extack, 1829 }; 1830 err = htb_offload(dev, &offload_opt); 1831 if (err) { 1832 pr_err("htb: TC_HTB_LEAF_ALLOC_QUEUE failed with err = %d\n", 1833 err); 1834 goto err_kill_estimator; 1835 } 1836 dev_queue = netdev_get_tx_queue(dev, offload_opt.qid); 1837 } else { /* First child. */ 1838 dev_queue = parent->leaf.q->dev_queue; 1839 old_q = htb_graft_helper(dev_queue, NULL); 1840 WARN_ON(old_q != parent->leaf.q); 1841 offload_opt = (struct tc_htb_qopt_offload) { 1842 .command = TC_HTB_LEAF_TO_INNER, 1843 .classid = cl->common.classid, 1844 .parent_classid = 1845 TC_H_MIN(parent->common.classid), 1846 .rate = max_t(u64, hopt->rate.rate, rate64), 1847 .ceil = max_t(u64, hopt->ceil.rate, ceil64), 1848 .extack = extack, 1849 }; 1850 err = htb_offload(dev, &offload_opt); 1851 if (err) { 1852 pr_err("htb: TC_HTB_LEAF_TO_INNER failed with err = %d\n", 1853 err); 1854 htb_graft_helper(dev_queue, old_q); 1855 goto err_kill_estimator; 1856 } 1857 parent->bstats_bias.bytes += old_q->bstats.bytes; 1858 parent->bstats_bias.packets += old_q->bstats.packets; 1859 qdisc_put(old_q); 1860 } 1861 new_q = qdisc_create_dflt(dev_queue, &pfifo_qdisc_ops, 1862 classid, NULL); 1863 if (q->offload) { 1864 if (new_q) { 1865 htb_set_lockdep_class_child(new_q); 1866 /* One ref for cl->leaf.q, the other for 1867 * dev_queue->qdisc. 1868 */ 1869 qdisc_refcount_inc(new_q); 1870 } 1871 old_q = htb_graft_helper(dev_queue, new_q); 1872 /* No qdisc_put needed. */ 1873 WARN_ON(!(old_q->flags & TCQ_F_BUILTIN)); 1874 } 1875 sch_tree_lock(sch); 1876 if (parent && !parent->level) { 1877 /* turn parent into inner node */ 1878 qdisc_purge_queue(parent->leaf.q); 1879 parent_qdisc = parent->leaf.q; 1880 if (parent->prio_activity) 1881 htb_deactivate(q, parent); 1882 1883 /* remove from evt list because of level change */ 1884 if (parent->cmode != HTB_CAN_SEND) { 1885 htb_safe_rb_erase(&parent->pq_node, &q->hlevel[0].wait_pq); 1886 parent->cmode = HTB_CAN_SEND; 1887 } 1888 parent->level = (parent->parent ? parent->parent->level 1889 : TC_HTB_MAXDEPTH) - 1; 1890 memset(&parent->inner, 0, sizeof(parent->inner)); 1891 } 1892 1893 /* leaf (we) needs elementary qdisc */ 1894 cl->leaf.q = new_q ? new_q : &noop_qdisc; 1895 1896 cl->parent = parent; 1897 1898 /* set class to be in HTB_CAN_SEND state */ 1899 cl->tokens = PSCHED_TICKS2NS(hopt->buffer); 1900 cl->ctokens = PSCHED_TICKS2NS(hopt->cbuffer); 1901 cl->mbuffer = 60ULL * NSEC_PER_SEC; /* 1min */ 1902 cl->t_c = ktime_get_ns(); 1903 cl->cmode = HTB_CAN_SEND; 1904 1905 /* attach to the hash list and parent's family */ 1906 qdisc_class_hash_insert(&q->clhash, &cl->common); 1907 if (parent) 1908 parent->children++; 1909 if (cl->leaf.q != &noop_qdisc) 1910 qdisc_hash_add(cl->leaf.q, true); 1911 } else { 1912 if (tca[TCA_RATE]) { 1913 err = gen_replace_estimator(&cl->bstats, NULL, 1914 &cl->rate_est, 1915 NULL, 1916 qdisc_root_sleeping_running(sch), 1917 tca[TCA_RATE]); 1918 if (err) 1919 return err; 1920 } 1921 1922 if (q->offload) { 1923 struct net_device *dev = qdisc_dev(sch); 1924 1925 offload_opt = (struct tc_htb_qopt_offload) { 1926 .command = TC_HTB_NODE_MODIFY, 1927 .classid = cl->common.classid, 1928 .rate = max_t(u64, hopt->rate.rate, rate64), 1929 .ceil = max_t(u64, hopt->ceil.rate, ceil64), 1930 .extack = extack, 1931 }; 1932 err = htb_offload(dev, &offload_opt); 1933 if (err) 1934 /* Estimator was replaced, and rollback may fail 1935 * as well, so we don't try to recover it, and 1936 * the estimator won't work property with the 1937 * offload anyway, because bstats are updated 1938 * only when the stats are queried. 1939 */ 1940 return err; 1941 } 1942 1943 sch_tree_lock(sch); 1944 } 1945 1946 psched_ratecfg_precompute(&cl->rate, &hopt->rate, rate64); 1947 psched_ratecfg_precompute(&cl->ceil, &hopt->ceil, ceil64); 1948 1949 /* it used to be a nasty bug here, we have to check that node 1950 * is really leaf before changing cl->leaf ! 1951 */ 1952 if (!cl->level) { 1953 u64 quantum = cl->rate.rate_bytes_ps; 1954 1955 do_div(quantum, q->rate2quantum); 1956 cl->quantum = min_t(u64, quantum, INT_MAX); 1957 1958 if (!hopt->quantum && cl->quantum < 1000) { 1959 warn = -1; 1960 cl->quantum = 1000; 1961 } 1962 if (!hopt->quantum && cl->quantum > 200000) { 1963 warn = 1; 1964 cl->quantum = 200000; 1965 } 1966 if (hopt->quantum) 1967 cl->quantum = hopt->quantum; 1968 if ((cl->prio = hopt->prio) >= TC_HTB_NUMPRIO) 1969 cl->prio = TC_HTB_NUMPRIO - 1; 1970 } 1971 1972 cl->buffer = PSCHED_TICKS2NS(hopt->buffer); 1973 cl->cbuffer = PSCHED_TICKS2NS(hopt->cbuffer); 1974 1975 sch_tree_unlock(sch); 1976 qdisc_put(parent_qdisc); 1977 1978 if (warn) 1979 pr_warn("HTB: quantum of class %X is %s. Consider r2q change.\n", 1980 cl->common.classid, (warn == -1 ? "small" : "big")); 1981 1982 qdisc_class_hash_grow(sch, &q->clhash); 1983 1984 *arg = (unsigned long)cl; 1985 return 0; 1986 1987 err_kill_estimator: 1988 gen_kill_estimator(&cl->rate_est); 1989 err_block_put: 1990 tcf_block_put(cl->block); 1991 kfree(cl); 1992 failure: 1993 return err; 1994 } 1995 1996 static struct tcf_block *htb_tcf_block(struct Qdisc *sch, unsigned long arg, 1997 struct netlink_ext_ack *extack) 1998 { 1999 struct htb_sched *q = qdisc_priv(sch); 2000 struct htb_class *cl = (struct htb_class *)arg; 2001 2002 return cl ? cl->block : q->block; 2003 } 2004 2005 static unsigned long htb_bind_filter(struct Qdisc *sch, unsigned long parent, 2006 u32 classid) 2007 { 2008 struct htb_class *cl = htb_find(classid, sch); 2009 2010 /*if (cl && !cl->level) return 0; 2011 * The line above used to be there to prevent attaching filters to 2012 * leaves. But at least tc_index filter uses this just to get class 2013 * for other reasons so that we have to allow for it. 2014 * ---- 2015 * 19.6.2002 As Werner explained it is ok - bind filter is just 2016 * another way to "lock" the class - unlike "get" this lock can 2017 * be broken by class during destroy IIUC. 2018 */ 2019 if (cl) 2020 cl->filter_cnt++; 2021 return (unsigned long)cl; 2022 } 2023 2024 static void htb_unbind_filter(struct Qdisc *sch, unsigned long arg) 2025 { 2026 struct htb_class *cl = (struct htb_class *)arg; 2027 2028 if (cl) 2029 cl->filter_cnt--; 2030 } 2031 2032 static void htb_walk(struct Qdisc *sch, struct qdisc_walker *arg) 2033 { 2034 struct htb_sched *q = qdisc_priv(sch); 2035 struct htb_class *cl; 2036 unsigned int i; 2037 2038 if (arg->stop) 2039 return; 2040 2041 for (i = 0; i < q->clhash.hashsize; i++) { 2042 hlist_for_each_entry(cl, &q->clhash.hash[i], common.hnode) { 2043 if (arg->count < arg->skip) { 2044 arg->count++; 2045 continue; 2046 } 2047 if (arg->fn(sch, (unsigned long)cl, arg) < 0) { 2048 arg->stop = 1; 2049 return; 2050 } 2051 arg->count++; 2052 } 2053 } 2054 } 2055 2056 static const struct Qdisc_class_ops htb_class_ops = { 2057 .select_queue = htb_select_queue, 2058 .graft = htb_graft, 2059 .leaf = htb_leaf, 2060 .qlen_notify = htb_qlen_notify, 2061 .find = htb_search, 2062 .change = htb_change_class, 2063 .delete = htb_delete, 2064 .walk = htb_walk, 2065 .tcf_block = htb_tcf_block, 2066 .bind_tcf = htb_bind_filter, 2067 .unbind_tcf = htb_unbind_filter, 2068 .dump = htb_dump_class, 2069 .dump_stats = htb_dump_class_stats, 2070 }; 2071 2072 static struct Qdisc_ops htb_qdisc_ops __read_mostly = { 2073 .cl_ops = &htb_class_ops, 2074 .id = "htb", 2075 .priv_size = sizeof(struct htb_sched), 2076 .enqueue = htb_enqueue, 2077 .dequeue = htb_dequeue, 2078 .peek = qdisc_peek_dequeued, 2079 .init = htb_init, 2080 .attach = htb_attach, 2081 .reset = htb_reset, 2082 .destroy = htb_destroy, 2083 .dump = htb_dump, 2084 .owner = THIS_MODULE, 2085 }; 2086 2087 static int __init htb_module_init(void) 2088 { 2089 return register_qdisc(&htb_qdisc_ops); 2090 } 2091 static void __exit htb_module_exit(void) 2092 { 2093 unregister_qdisc(&htb_qdisc_ops); 2094 } 2095 2096 module_init(htb_module_init) 2097 module_exit(htb_module_exit) 2098 MODULE_LICENSE("GPL"); 2099