1 /* 2 * Fast and scalable bitmap tagging variant. Uses sparser bitmaps spread 3 * over multiple cachelines to avoid ping-pong between multiple submitters 4 * or submitter and completer. Uses rolling wakeups to avoid falling of 5 * the scaling cliff when we run out of tags and have to start putting 6 * submitters to sleep. 7 * 8 * Uses active queue tracking to support fairer distribution of tags 9 * between multiple submitters when a shared tag map is used. 10 * 11 * Copyright (C) 2013-2014 Jens Axboe 12 */ 13 #include <linux/kernel.h> 14 #include <linux/module.h> 15 #include <linux/random.h> 16 17 #include <linux/blk-mq.h> 18 #include "blk.h" 19 #include "blk-mq.h" 20 #include "blk-mq-tag.h" 21 22 static bool bt_has_free_tags(struct blk_mq_bitmap_tags *bt) 23 { 24 int i; 25 26 for (i = 0; i < bt->map_nr; i++) { 27 struct blk_align_bitmap *bm = &bt->map[i]; 28 int ret; 29 30 ret = find_first_zero_bit(&bm->word, bm->depth); 31 if (ret < bm->depth) 32 return true; 33 } 34 35 return false; 36 } 37 38 bool blk_mq_has_free_tags(struct blk_mq_tags *tags) 39 { 40 if (!tags) 41 return true; 42 43 return bt_has_free_tags(&tags->bitmap_tags); 44 } 45 46 static inline int bt_index_inc(int index) 47 { 48 return (index + 1) & (BT_WAIT_QUEUES - 1); 49 } 50 51 static inline void bt_index_atomic_inc(atomic_t *index) 52 { 53 int old = atomic_read(index); 54 int new = bt_index_inc(old); 55 atomic_cmpxchg(index, old, new); 56 } 57 58 /* 59 * If a previously inactive queue goes active, bump the active user count. 60 */ 61 bool __blk_mq_tag_busy(struct blk_mq_hw_ctx *hctx) 62 { 63 if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state) && 64 !test_and_set_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) 65 atomic_inc(&hctx->tags->active_queues); 66 67 return true; 68 } 69 70 /* 71 * Wakeup all potentially sleeping on tags 72 */ 73 void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool include_reserve) 74 { 75 struct blk_mq_bitmap_tags *bt; 76 int i, wake_index; 77 78 /* 79 * Make sure all changes prior to this are visible from other CPUs. 80 */ 81 smp_mb(); 82 bt = &tags->bitmap_tags; 83 wake_index = atomic_read(&bt->wake_index); 84 for (i = 0; i < BT_WAIT_QUEUES; i++) { 85 struct bt_wait_state *bs = &bt->bs[wake_index]; 86 87 if (waitqueue_active(&bs->wait)) 88 wake_up(&bs->wait); 89 90 wake_index = bt_index_inc(wake_index); 91 } 92 93 if (include_reserve) { 94 bt = &tags->breserved_tags; 95 if (waitqueue_active(&bt->bs[0].wait)) 96 wake_up(&bt->bs[0].wait); 97 } 98 } 99 100 /* 101 * If a previously busy queue goes inactive, potential waiters could now 102 * be allowed to queue. Wake them up and check. 103 */ 104 void __blk_mq_tag_idle(struct blk_mq_hw_ctx *hctx) 105 { 106 struct blk_mq_tags *tags = hctx->tags; 107 108 if (!test_and_clear_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) 109 return; 110 111 atomic_dec(&tags->active_queues); 112 113 blk_mq_tag_wakeup_all(tags, false); 114 } 115 116 /* 117 * For shared tag users, we track the number of currently active users 118 * and attempt to provide a fair share of the tag depth for each of them. 119 */ 120 static inline bool hctx_may_queue(struct blk_mq_hw_ctx *hctx, 121 struct blk_mq_bitmap_tags *bt) 122 { 123 unsigned int depth, users; 124 125 if (!hctx || !(hctx->flags & BLK_MQ_F_TAG_SHARED)) 126 return true; 127 if (!test_bit(BLK_MQ_S_TAG_ACTIVE, &hctx->state)) 128 return true; 129 130 /* 131 * Don't try dividing an ant 132 */ 133 if (bt->depth == 1) 134 return true; 135 136 users = atomic_read(&hctx->tags->active_queues); 137 if (!users) 138 return true; 139 140 /* 141 * Allow at least some tags 142 */ 143 depth = max((bt->depth + users - 1) / users, 4U); 144 return atomic_read(&hctx->nr_active) < depth; 145 } 146 147 static int __bt_get_word(struct blk_align_bitmap *bm, unsigned int last_tag, 148 bool nowrap) 149 { 150 int tag, org_last_tag = last_tag; 151 152 while (1) { 153 tag = find_next_zero_bit(&bm->word, bm->depth, last_tag); 154 if (unlikely(tag >= bm->depth)) { 155 /* 156 * We started with an offset, and we didn't reset the 157 * offset to 0 in a failure case, so start from 0 to 158 * exhaust the map. 159 */ 160 if (org_last_tag && last_tag && !nowrap) { 161 last_tag = org_last_tag = 0; 162 continue; 163 } 164 return -1; 165 } 166 167 if (!test_and_set_bit(tag, &bm->word)) 168 break; 169 170 last_tag = tag + 1; 171 if (last_tag >= bm->depth - 1) 172 last_tag = 0; 173 } 174 175 return tag; 176 } 177 178 #define BT_ALLOC_RR(tags) (tags->alloc_policy == BLK_TAG_ALLOC_RR) 179 180 /* 181 * Straight forward bitmap tag implementation, where each bit is a tag 182 * (cleared == free, and set == busy). The small twist is using per-cpu 183 * last_tag caches, which blk-mq stores in the blk_mq_ctx software queue 184 * contexts. This enables us to drastically limit the space searched, 185 * without dirtying an extra shared cacheline like we would if we stored 186 * the cache value inside the shared blk_mq_bitmap_tags structure. On top 187 * of that, each word of tags is in a separate cacheline. This means that 188 * multiple users will tend to stick to different cachelines, at least 189 * until the map is exhausted. 190 */ 191 static int __bt_get(struct blk_mq_hw_ctx *hctx, struct blk_mq_bitmap_tags *bt, 192 unsigned int *tag_cache, struct blk_mq_tags *tags) 193 { 194 unsigned int last_tag, org_last_tag; 195 int index, i, tag; 196 197 if (!hctx_may_queue(hctx, bt)) 198 return -1; 199 200 last_tag = org_last_tag = *tag_cache; 201 index = TAG_TO_INDEX(bt, last_tag); 202 203 for (i = 0; i < bt->map_nr; i++) { 204 tag = __bt_get_word(&bt->map[index], TAG_TO_BIT(bt, last_tag), 205 BT_ALLOC_RR(tags)); 206 if (tag != -1) { 207 tag += (index << bt->bits_per_word); 208 goto done; 209 } 210 211 /* 212 * Jump to next index, and reset the last tag to be the 213 * first tag of that index 214 */ 215 index++; 216 last_tag = (index << bt->bits_per_word); 217 218 if (index >= bt->map_nr) { 219 index = 0; 220 last_tag = 0; 221 } 222 } 223 224 *tag_cache = 0; 225 return -1; 226 227 /* 228 * Only update the cache from the allocation path, if we ended 229 * up using the specific cached tag. 230 */ 231 done: 232 if (tag == org_last_tag || unlikely(BT_ALLOC_RR(tags))) { 233 last_tag = tag + 1; 234 if (last_tag >= bt->depth - 1) 235 last_tag = 0; 236 237 *tag_cache = last_tag; 238 } 239 240 return tag; 241 } 242 243 static struct bt_wait_state *bt_wait_ptr(struct blk_mq_bitmap_tags *bt, 244 struct blk_mq_hw_ctx *hctx) 245 { 246 struct bt_wait_state *bs; 247 int wait_index; 248 249 if (!hctx) 250 return &bt->bs[0]; 251 252 wait_index = atomic_read(&hctx->wait_index); 253 bs = &bt->bs[wait_index]; 254 bt_index_atomic_inc(&hctx->wait_index); 255 return bs; 256 } 257 258 static int bt_get(struct blk_mq_alloc_data *data, 259 struct blk_mq_bitmap_tags *bt, 260 struct blk_mq_hw_ctx *hctx, 261 unsigned int *last_tag, struct blk_mq_tags *tags) 262 { 263 struct bt_wait_state *bs; 264 DEFINE_WAIT(wait); 265 int tag; 266 267 tag = __bt_get(hctx, bt, last_tag, tags); 268 if (tag != -1) 269 return tag; 270 271 if (data->flags & BLK_MQ_REQ_NOWAIT) 272 return -1; 273 274 bs = bt_wait_ptr(bt, hctx); 275 do { 276 prepare_to_wait(&bs->wait, &wait, TASK_UNINTERRUPTIBLE); 277 278 tag = __bt_get(hctx, bt, last_tag, tags); 279 if (tag != -1) 280 break; 281 282 /* 283 * We're out of tags on this hardware queue, kick any 284 * pending IO submits before going to sleep waiting for 285 * some to complete. Note that hctx can be NULL here for 286 * reserved tag allocation. 287 */ 288 if (hctx) 289 blk_mq_run_hw_queue(hctx, false); 290 291 /* 292 * Retry tag allocation after running the hardware queue, 293 * as running the queue may also have found completions. 294 */ 295 tag = __bt_get(hctx, bt, last_tag, tags); 296 if (tag != -1) 297 break; 298 299 blk_mq_put_ctx(data->ctx); 300 301 io_schedule(); 302 303 data->ctx = blk_mq_get_ctx(data->q); 304 data->hctx = data->q->mq_ops->map_queue(data->q, 305 data->ctx->cpu); 306 if (data->flags & BLK_MQ_REQ_RESERVED) { 307 bt = &data->hctx->tags->breserved_tags; 308 } else { 309 last_tag = &data->ctx->last_tag; 310 hctx = data->hctx; 311 bt = &hctx->tags->bitmap_tags; 312 } 313 finish_wait(&bs->wait, &wait); 314 bs = bt_wait_ptr(bt, hctx); 315 } while (1); 316 317 finish_wait(&bs->wait, &wait); 318 return tag; 319 } 320 321 static unsigned int __blk_mq_get_tag(struct blk_mq_alloc_data *data) 322 { 323 int tag; 324 325 tag = bt_get(data, &data->hctx->tags->bitmap_tags, data->hctx, 326 &data->ctx->last_tag, data->hctx->tags); 327 if (tag >= 0) 328 return tag + data->hctx->tags->nr_reserved_tags; 329 330 return BLK_MQ_TAG_FAIL; 331 } 332 333 static unsigned int __blk_mq_get_reserved_tag(struct blk_mq_alloc_data *data) 334 { 335 int tag, zero = 0; 336 337 if (unlikely(!data->hctx->tags->nr_reserved_tags)) { 338 WARN_ON_ONCE(1); 339 return BLK_MQ_TAG_FAIL; 340 } 341 342 tag = bt_get(data, &data->hctx->tags->breserved_tags, NULL, &zero, 343 data->hctx->tags); 344 if (tag < 0) 345 return BLK_MQ_TAG_FAIL; 346 347 return tag; 348 } 349 350 unsigned int blk_mq_get_tag(struct blk_mq_alloc_data *data) 351 { 352 if (data->flags & BLK_MQ_REQ_RESERVED) 353 return __blk_mq_get_reserved_tag(data); 354 return __blk_mq_get_tag(data); 355 } 356 357 static struct bt_wait_state *bt_wake_ptr(struct blk_mq_bitmap_tags *bt) 358 { 359 int i, wake_index; 360 361 wake_index = atomic_read(&bt->wake_index); 362 for (i = 0; i < BT_WAIT_QUEUES; i++) { 363 struct bt_wait_state *bs = &bt->bs[wake_index]; 364 365 if (waitqueue_active(&bs->wait)) { 366 int o = atomic_read(&bt->wake_index); 367 if (wake_index != o) 368 atomic_cmpxchg(&bt->wake_index, o, wake_index); 369 370 return bs; 371 } 372 373 wake_index = bt_index_inc(wake_index); 374 } 375 376 return NULL; 377 } 378 379 static void bt_clear_tag(struct blk_mq_bitmap_tags *bt, unsigned int tag) 380 { 381 const int index = TAG_TO_INDEX(bt, tag); 382 struct bt_wait_state *bs; 383 int wait_cnt; 384 385 clear_bit(TAG_TO_BIT(bt, tag), &bt->map[index].word); 386 387 /* Ensure that the wait list checks occur after clear_bit(). */ 388 smp_mb(); 389 390 bs = bt_wake_ptr(bt); 391 if (!bs) 392 return; 393 394 wait_cnt = atomic_dec_return(&bs->wait_cnt); 395 if (unlikely(wait_cnt < 0)) 396 wait_cnt = atomic_inc_return(&bs->wait_cnt); 397 if (wait_cnt == 0) { 398 atomic_add(bt->wake_cnt, &bs->wait_cnt); 399 bt_index_atomic_inc(&bt->wake_index); 400 wake_up(&bs->wait); 401 } 402 } 403 404 void blk_mq_put_tag(struct blk_mq_hw_ctx *hctx, unsigned int tag, 405 unsigned int *last_tag) 406 { 407 struct blk_mq_tags *tags = hctx->tags; 408 409 if (tag >= tags->nr_reserved_tags) { 410 const int real_tag = tag - tags->nr_reserved_tags; 411 412 BUG_ON(real_tag >= tags->nr_tags); 413 bt_clear_tag(&tags->bitmap_tags, real_tag); 414 if (likely(tags->alloc_policy == BLK_TAG_ALLOC_FIFO)) 415 *last_tag = real_tag; 416 } else { 417 BUG_ON(tag >= tags->nr_reserved_tags); 418 bt_clear_tag(&tags->breserved_tags, tag); 419 } 420 } 421 422 static void bt_for_each(struct blk_mq_hw_ctx *hctx, 423 struct blk_mq_bitmap_tags *bt, unsigned int off, 424 busy_iter_fn *fn, void *data, bool reserved) 425 { 426 struct request *rq; 427 int bit, i; 428 429 for (i = 0; i < bt->map_nr; i++) { 430 struct blk_align_bitmap *bm = &bt->map[i]; 431 432 for (bit = find_first_bit(&bm->word, bm->depth); 433 bit < bm->depth; 434 bit = find_next_bit(&bm->word, bm->depth, bit + 1)) { 435 rq = hctx->tags->rqs[off + bit]; 436 if (rq->q == hctx->queue) 437 fn(hctx, rq, data, reserved); 438 } 439 440 off += (1 << bt->bits_per_word); 441 } 442 } 443 444 static void bt_tags_for_each(struct blk_mq_tags *tags, 445 struct blk_mq_bitmap_tags *bt, unsigned int off, 446 busy_tag_iter_fn *fn, void *data, bool reserved) 447 { 448 struct request *rq; 449 int bit, i; 450 451 if (!tags->rqs) 452 return; 453 for (i = 0; i < bt->map_nr; i++) { 454 struct blk_align_bitmap *bm = &bt->map[i]; 455 456 for (bit = find_first_bit(&bm->word, bm->depth); 457 bit < bm->depth; 458 bit = find_next_bit(&bm->word, bm->depth, bit + 1)) { 459 rq = tags->rqs[off + bit]; 460 fn(rq, data, reserved); 461 } 462 463 off += (1 << bt->bits_per_word); 464 } 465 } 466 467 void blk_mq_all_tag_busy_iter(struct blk_mq_tags *tags, busy_tag_iter_fn *fn, 468 void *priv) 469 { 470 if (tags->nr_reserved_tags) 471 bt_tags_for_each(tags, &tags->breserved_tags, 0, fn, priv, true); 472 bt_tags_for_each(tags, &tags->bitmap_tags, tags->nr_reserved_tags, fn, priv, 473 false); 474 } 475 EXPORT_SYMBOL(blk_mq_all_tag_busy_iter); 476 477 void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, 478 void *priv) 479 { 480 struct blk_mq_hw_ctx *hctx; 481 int i; 482 483 484 queue_for_each_hw_ctx(q, hctx, i) { 485 struct blk_mq_tags *tags = hctx->tags; 486 487 /* 488 * If not software queues are currently mapped to this 489 * hardware queue, there's nothing to check 490 */ 491 if (!blk_mq_hw_queue_mapped(hctx)) 492 continue; 493 494 if (tags->nr_reserved_tags) 495 bt_for_each(hctx, &tags->breserved_tags, 0, fn, priv, true); 496 bt_for_each(hctx, &tags->bitmap_tags, tags->nr_reserved_tags, fn, priv, 497 false); 498 } 499 500 } 501 502 static unsigned int bt_unused_tags(struct blk_mq_bitmap_tags *bt) 503 { 504 unsigned int i, used; 505 506 for (i = 0, used = 0; i < bt->map_nr; i++) { 507 struct blk_align_bitmap *bm = &bt->map[i]; 508 509 used += bitmap_weight(&bm->word, bm->depth); 510 } 511 512 return bt->depth - used; 513 } 514 515 static void bt_update_count(struct blk_mq_bitmap_tags *bt, 516 unsigned int depth) 517 { 518 unsigned int tags_per_word = 1U << bt->bits_per_word; 519 unsigned int map_depth = depth; 520 521 if (depth) { 522 int i; 523 524 for (i = 0; i < bt->map_nr; i++) { 525 bt->map[i].depth = min(map_depth, tags_per_word); 526 map_depth -= bt->map[i].depth; 527 } 528 } 529 530 bt->wake_cnt = BT_WAIT_BATCH; 531 if (bt->wake_cnt > depth / BT_WAIT_QUEUES) 532 bt->wake_cnt = max(1U, depth / BT_WAIT_QUEUES); 533 534 bt->depth = depth; 535 } 536 537 static int bt_alloc(struct blk_mq_bitmap_tags *bt, unsigned int depth, 538 int node, bool reserved) 539 { 540 int i; 541 542 bt->bits_per_word = ilog2(BITS_PER_LONG); 543 544 /* 545 * Depth can be zero for reserved tags, that's not a failure 546 * condition. 547 */ 548 if (depth) { 549 unsigned int nr, tags_per_word; 550 551 tags_per_word = (1 << bt->bits_per_word); 552 553 /* 554 * If the tag space is small, shrink the number of tags 555 * per word so we spread over a few cachelines, at least. 556 * If less than 4 tags, just forget about it, it's not 557 * going to work optimally anyway. 558 */ 559 if (depth >= 4) { 560 while (tags_per_word * 4 > depth) { 561 bt->bits_per_word--; 562 tags_per_word = (1 << bt->bits_per_word); 563 } 564 } 565 566 nr = ALIGN(depth, tags_per_word) / tags_per_word; 567 bt->map = kzalloc_node(nr * sizeof(struct blk_align_bitmap), 568 GFP_KERNEL, node); 569 if (!bt->map) 570 return -ENOMEM; 571 572 bt->map_nr = nr; 573 } 574 575 bt->bs = kzalloc(BT_WAIT_QUEUES * sizeof(*bt->bs), GFP_KERNEL); 576 if (!bt->bs) { 577 kfree(bt->map); 578 bt->map = NULL; 579 return -ENOMEM; 580 } 581 582 bt_update_count(bt, depth); 583 584 for (i = 0; i < BT_WAIT_QUEUES; i++) { 585 init_waitqueue_head(&bt->bs[i].wait); 586 atomic_set(&bt->bs[i].wait_cnt, bt->wake_cnt); 587 } 588 589 return 0; 590 } 591 592 static void bt_free(struct blk_mq_bitmap_tags *bt) 593 { 594 kfree(bt->map); 595 kfree(bt->bs); 596 } 597 598 static struct blk_mq_tags *blk_mq_init_bitmap_tags(struct blk_mq_tags *tags, 599 int node, int alloc_policy) 600 { 601 unsigned int depth = tags->nr_tags - tags->nr_reserved_tags; 602 603 tags->alloc_policy = alloc_policy; 604 605 if (bt_alloc(&tags->bitmap_tags, depth, node, false)) 606 goto enomem; 607 if (bt_alloc(&tags->breserved_tags, tags->nr_reserved_tags, node, true)) 608 goto enomem; 609 610 return tags; 611 enomem: 612 bt_free(&tags->bitmap_tags); 613 kfree(tags); 614 return NULL; 615 } 616 617 struct blk_mq_tags *blk_mq_init_tags(unsigned int total_tags, 618 unsigned int reserved_tags, 619 int node, int alloc_policy) 620 { 621 struct blk_mq_tags *tags; 622 623 if (total_tags > BLK_MQ_TAG_MAX) { 624 pr_err("blk-mq: tag depth too large\n"); 625 return NULL; 626 } 627 628 tags = kzalloc_node(sizeof(*tags), GFP_KERNEL, node); 629 if (!tags) 630 return NULL; 631 632 if (!zalloc_cpumask_var(&tags->cpumask, GFP_KERNEL)) { 633 kfree(tags); 634 return NULL; 635 } 636 637 tags->nr_tags = total_tags; 638 tags->nr_reserved_tags = reserved_tags; 639 640 return blk_mq_init_bitmap_tags(tags, node, alloc_policy); 641 } 642 643 void blk_mq_free_tags(struct blk_mq_tags *tags) 644 { 645 bt_free(&tags->bitmap_tags); 646 bt_free(&tags->breserved_tags); 647 free_cpumask_var(tags->cpumask); 648 kfree(tags); 649 } 650 651 void blk_mq_tag_init_last_tag(struct blk_mq_tags *tags, unsigned int *tag) 652 { 653 unsigned int depth = tags->nr_tags - tags->nr_reserved_tags; 654 655 *tag = prandom_u32() % depth; 656 } 657 658 int blk_mq_tag_update_depth(struct blk_mq_tags *tags, unsigned int tdepth) 659 { 660 tdepth -= tags->nr_reserved_tags; 661 if (tdepth > tags->nr_tags) 662 return -EINVAL; 663 664 /* 665 * Don't need (or can't) update reserved tags here, they remain 666 * static and should never need resizing. 667 */ 668 bt_update_count(&tags->bitmap_tags, tdepth); 669 blk_mq_tag_wakeup_all(tags, false); 670 return 0; 671 } 672 673 /** 674 * blk_mq_unique_tag() - return a tag that is unique queue-wide 675 * @rq: request for which to compute a unique tag 676 * 677 * The tag field in struct request is unique per hardware queue but not over 678 * all hardware queues. Hence this function that returns a tag with the 679 * hardware context index in the upper bits and the per hardware queue tag in 680 * the lower bits. 681 * 682 * Note: When called for a request that is queued on a non-multiqueue request 683 * queue, the hardware context index is set to zero. 684 */ 685 u32 blk_mq_unique_tag(struct request *rq) 686 { 687 struct request_queue *q = rq->q; 688 struct blk_mq_hw_ctx *hctx; 689 int hwq = 0; 690 691 if (q->mq_ops) { 692 hctx = q->mq_ops->map_queue(q, rq->mq_ctx->cpu); 693 hwq = hctx->queue_num; 694 } 695 696 return (hwq << BLK_MQ_UNIQUE_TAG_BITS) | 697 (rq->tag & BLK_MQ_UNIQUE_TAG_MASK); 698 } 699 EXPORT_SYMBOL(blk_mq_unique_tag); 700 701 ssize_t blk_mq_tag_sysfs_show(struct blk_mq_tags *tags, char *page) 702 { 703 char *orig_page = page; 704 unsigned int free, res; 705 706 if (!tags) 707 return 0; 708 709 page += sprintf(page, "nr_tags=%u, reserved_tags=%u, " 710 "bits_per_word=%u\n", 711 tags->nr_tags, tags->nr_reserved_tags, 712 tags->bitmap_tags.bits_per_word); 713 714 free = bt_unused_tags(&tags->bitmap_tags); 715 res = bt_unused_tags(&tags->breserved_tags); 716 717 page += sprintf(page, "nr_free=%u, nr_reserved=%u\n", free, res); 718 page += sprintf(page, "active_queues=%u\n", atomic_read(&tags->active_queues)); 719 720 return page - orig_page; 721 } 722