summaryrefslogtreecommitdiff
path: root/block/blk-wbt.c
blob: 8cb53bf4c7b152de218684d334409521dabf9178 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
// SPDX-License-Identifier: GPL-2.0
/*
 * buffered writeback throttling. loosely based on CoDel. We can't drop
 * packets for IO scheduling, so the logic is something like this:
 *
 * - Monitor latencies in a defined window of time.
 * - If the minimum latency in the above window exceeds some target, increment
 *   scaling step and scale down queue depth by a factor of 2x. The monitoring
 *   window is then shrunk to 100 / sqrt(scaling step + 1).
 * - For any window where we don't have solid data on what the latencies
 *   look like, retain status quo.
 * - If latencies look good, decrement scaling step.
 * - If we're only doing writes, allow the scaling step to go negative. This
 *   will temporarily boost write performance, snapping back to a stable
 *   scaling step of 0 if reads show up or the heavy writers finish. Unlike
 *   positive scaling steps where we shrink the monitoring window, a negative
 *   scaling step retains the default step==0 window size.
 *
 * Copyright (C) 2016 Jens Axboe
 *
 */
#include <linux/kernel.h>
#include <linux/blk_types.h>
#include <linux/slab.h>
#include <linux/backing-dev.h>
#include <linux/swap.h>

#include "blk-stat.h"
#include "blk-wbt.h"
#include "blk-rq-qos.h"
#include "elevator.h"
#include "blk.h"

#define CREATE_TRACE_POINTS
#include <trace/events/wbt.h>

enum wbt_flags {
	WBT_TRACKED		= 1,	/* write, tracked for throttling */
	WBT_READ		= 2,	/* read */
	WBT_KSWAPD		= 4,	/* write, from kswapd */
	WBT_DISCARD		= 8,	/* discard */

	WBT_NR_BITS		= 4,	/* number of bits */
};

enum {
	WBT_RWQ_BG		= 0,
	WBT_RWQ_KSWAPD,
	WBT_RWQ_DISCARD,
	WBT_NUM_RWQ,
};

/*
 * If current state is WBT_STATE_ON/OFF_DEFAULT, it can be covered to any other
 * state, if current state is WBT_STATE_ON/OFF_MANUAL, it can only be covered
 * to WBT_STATE_OFF/ON_MANUAL.
 */
enum {
	WBT_STATE_ON_DEFAULT	= 1,	/* on by default */
	WBT_STATE_ON_MANUAL	= 2,	/* on manually by sysfs */
	WBT_STATE_OFF_DEFAULT	= 3,	/* off by default */
	WBT_STATE_OFF_MANUAL	= 4,	/* off manually by sysfs */
};

struct rq_wb {
	/*
	 * Settings that govern how we throttle
	 */
	unsigned int wb_background;		/* background writeback */
	unsigned int wb_normal;			/* normal writeback */

	short enable_state;			/* WBT_STATE_* */

	/*
	 * Number of consecutive periods where we don't have enough
	 * information to make a firm scale up/down decision.
	 */
	unsigned int unknown_cnt;

	u64 win_nsec;				/* default window size */
	u64 cur_win_nsec;			/* current window size */

	struct blk_stat_callback *cb;

	u64 sync_issue;
	void *sync_cookie;

	unsigned long last_issue;		/* last non-throttled issue */
	unsigned long last_comp;		/* last non-throttled comp */
	unsigned long min_lat_nsec;
	struct rq_qos rqos;
	struct rq_wait rq_wait[WBT_NUM_RWQ];
	struct rq_depth rq_depth;
};

static inline struct rq_wb *RQWB(struct rq_qos *rqos)
{
	return container_of(rqos, struct rq_wb, rqos);
}

static inline void wbt_clear_state(struct request *rq)
{
	rq->wbt_flags = 0;
}

static inline enum wbt_flags wbt_flags(struct request *rq)
{
	return rq->wbt_flags;
}

static inline bool wbt_is_tracked(struct request *rq)
{
	return rq->wbt_flags & WBT_TRACKED;
}

static inline bool wbt_is_read(struct request *rq)
{
	return rq->wbt_flags & WBT_READ;
}

enum {
	/*
	 * Default setting, we'll scale up (to 75% of QD max) or down (min 1)
	 * from here depending on device stats
	 */
	RWB_DEF_DEPTH	= 16,

	/*
	 * 100msec window
	 */
	RWB_WINDOW_NSEC		= 100 * 1000 * 1000ULL,

	/*
	 * Disregard stats, if we don't meet this minimum
	 */
	RWB_MIN_WRITE_SAMPLES	= 3,

	/*
	 * If we have this number of consecutive windows with not enough
	 * information to scale up or down, scale up.
	 */
	RWB_UNKNOWN_BUMP	= 5,
};

static inline bool rwb_enabled(struct rq_wb *rwb)
{
	return rwb && rwb->enable_state != WBT_STATE_OFF_DEFAULT &&
		      rwb->enable_state != WBT_STATE_OFF_MANUAL;
}

static void wb_timestamp(struct rq_wb *rwb, unsigned long *var)
{
	if (rwb_enabled(rwb)) {
		const unsigned long cur = jiffies;

		if (cur != *var)
			*var = cur;
	}
}

/*
 * If a task was rate throttled in balance_dirty_pages() within the last
 * second or so, use that to indicate a higher cleaning rate.
 */
static bool wb_recent_wait(struct rq_wb *rwb)
{
	struct bdi_writeback *wb = &rwb->rqos.disk->bdi->wb;

	return time_before(jiffies, wb->dirty_sleep + HZ);
}

static inline struct rq_wait *get_rq_wait(struct rq_wb *rwb,
					  enum wbt_flags wb_acct)
{
	if (wb_acct & WBT_KSWAPD)
		return &rwb->rq_wait[WBT_RWQ_KSWAPD];
	else if (wb_acct & WBT_DISCARD)
		return &rwb->rq_wait[WBT_RWQ_DISCARD];

	return &rwb->rq_wait[WBT_RWQ_BG];
}

static void rwb_wake_all(struct rq_wb *rwb)
{
	int i;

	for (i = 0; i < WBT_NUM_RWQ; i++) {
		struct rq_wait *rqw = &rwb->rq_wait[i];

		if (wq_has_sleeper(&rqw->wait))
			wake_up_all(&rqw->wait);
	}
}

static void wbt_rqw_done(struct rq_wb *rwb, struct rq_wait *rqw,
			 enum wbt_flags wb_acct)
{
	int inflight, limit;

	inflight = atomic_dec_return(&rqw->inflight);

	/*
	 * For discards, our limit is always the background. For writes, if
	 * the device does write back caching, drop further down before we
	 * wake people up.
	 */
	if (wb_acct & WBT_DISCARD)
		limit = rwb->wb_background;
	else if (test_bit(QUEUE_FLAG_WC, &rwb->rqos.disk->queue->queue_flags) &&
	         !wb_recent_wait(rwb))
		limit = 0;
	else
		limit = rwb->wb_normal;

	/*
	 * Don't wake anyone up if we are above the normal limit.
	 */
	if (inflight && inflight >= limit)
		return;

	if (wq_has_sleeper(&rqw->wait)) {
		int diff = limit - inflight;

		if (!inflight || diff >= rwb->wb_background / 2)
			wake_up_all(&rqw->wait);
	}
}

static void __wbt_done(struct rq_qos *rqos, enum wbt_flags wb_acct)
{
	struct rq_wb *rwb = RQWB(rqos);
	struct rq_wait *rqw;

	if (!(wb_acct & WBT_TRACKED))
		return;

	rqw = get_rq_wait(rwb, wb_acct);
	wbt_rqw_done(rwb, rqw, wb_acct);
}

/*
 * Called on completion of a request. Note that it's also called when
 * a request is merged, when the request gets freed.
 */
static void wbt_done(struct rq_qos *rqos, struct request *rq)
{
	struct rq_wb *rwb = RQWB(rqos);

	if (!wbt_is_tracked(rq)) {
		if (rwb->sync_cookie == rq) {
			rwb->sync_issue = 0;
			rwb->sync_cookie = NULL;
		}

		if (wbt_is_read(rq))
			wb_timestamp(rwb, &rwb->last_comp);
	} else {
		WARN_ON_ONCE(rq == rwb->sync_cookie);
		__wbt_done(rqos, wbt_flags(rq));
	}
	wbt_clear_state(rq);
}

static inline bool stat_sample_valid(struct blk_rq_stat *stat)
{
	/*
	 * We need at least one read sample, and a minimum of
	 * RWB_MIN_WRITE_SAMPLES. We require some write samples to know
	 * that it's writes impacting us, and not just some sole read on
	 * a device that is in a lower power state.
	 */
	return (stat[READ].nr_samples >= 1 &&
		stat[WRITE].nr_samples >= RWB_MIN_WRITE_SAMPLES);
}

static u64 rwb_sync_issue_lat(struct rq_wb *rwb)
{
	u64 issue = READ_ONCE(rwb->sync_issue);

	if (!issue || !rwb->sync_cookie)
		return 0;

	return blk_time_get_ns() - issue;
}

static inline unsigned int wbt_inflight(struct rq_wb *rwb)
{
	unsigned int i, ret = 0;

	for (i = 0; i < WBT_NUM_RWQ; i++)
		ret += atomic_read(&rwb->rq_wait[i].inflight);

	return ret;
}

enum {
	LAT_OK = 1,
	LAT_UNKNOWN,
	LAT_UNKNOWN_WRITES,
	LAT_EXCEEDED,
};

static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat)
{
	struct backing_dev_info *bdi = rwb->rqos.disk->bdi;
	struct rq_depth *rqd = &rwb->rq_depth;
	u64 thislat;

	/*
	 * If our stored sync issue exceeds the window size, or it
	 * exceeds our min target AND we haven't logged any entries,
	 * flag the latency as exceeded. wbt works off completion latencies,
	 * but for a flooded device, a single sync IO can take a long time
	 * to complete after being issued. If this time exceeds our
	 * monitoring window AND we didn't see any other completions in that
	 * window, then count that sync IO as a violation of the latency.
	 */
	thislat = rwb_sync_issue_lat(rwb);
	if (thislat > rwb->cur_win_nsec ||
	    (thislat > rwb->min_lat_nsec && !stat[READ].nr_samples)) {
		trace_wbt_lat(bdi, thislat);
		return LAT_EXCEEDED;
	}

	/*
	 * No read/write mix, if stat isn't valid
	 */
	if (!stat_sample_valid(stat)) {
		/*
		 * If we had writes in this stat window and the window is
		 * current, we're only doing writes. If a task recently
		 * waited or still has writes in flights, consider us doing
		 * just writes as well.
		 */
		if (stat[WRITE].nr_samples || wb_recent_wait(rwb) ||
		    wbt_inflight(rwb))
			return LAT_UNKNOWN_WRITES;
		return LAT_UNKNOWN;
	}

	/*
	 * If the 'min' latency exceeds our target, step down.
	 */
	if (stat[READ].min > rwb->min_lat_nsec) {
		trace_wbt_lat(bdi, stat[READ].min);
		trace_wbt_stat(bdi, stat);
		return LAT_EXCEEDED;
	}

	if (rqd->scale_step)
		trace_wbt_stat(bdi, stat);

	return LAT_OK;
}

static void rwb_trace_step(struct rq_wb *rwb, const char *msg)
{
	struct backing_dev_info *bdi = rwb->rqos.disk->bdi;
	struct rq_depth *rqd = &rwb->rq_depth;

	trace_wbt_step(bdi, msg, rqd->scale_step, rwb->cur_win_nsec,
			rwb->wb_background, rwb->wb_normal, rqd->max_depth);
}

static void calc_wb_limits(struct rq_wb *rwb)
{
	if (rwb->min_lat_nsec == 0) {
		rwb->wb_normal = rwb->wb_background = 0;
	} else if (rwb->rq_depth.max_depth <= 2) {
		rwb->wb_normal = rwb->rq_depth.max_depth;
		rwb->wb_background = 1;
	} else {
		rwb->wb_normal = (rwb->rq_depth.max_depth + 1) / 2;
		rwb->wb_background = (rwb->rq_depth.max_depth + 3) / 4;
	}
}

static void scale_up(struct rq_wb *rwb)
{
	if (!rq_depth_scale_up(&rwb->rq_depth))
		return;
	calc_wb_limits(rwb);
	rwb->unknown_cnt = 0;
	rwb_wake_all(rwb);
	rwb_trace_step(rwb, tracepoint_string("scale up"));
}

static void scale_down(struct rq_wb *rwb, bool hard_throttle)
{
	if (!rq_depth_scale_down(&rwb->rq_depth, hard_throttle))
		return;
	calc_wb_limits(rwb);
	rwb->unknown_cnt = 0;
	rwb_trace_step(rwb, tracepoint_string("scale down"));
}

static void rwb_arm_timer(struct rq_wb *rwb)
{
	struct rq_depth *rqd = &rwb->rq_depth;

	if (rqd->scale_step > 0) {
		/*
		 * We should speed this up, using some variant of a fast
		 * integer inverse square root calculation. Since we only do
		 * this for every window expiration, it's not a huge deal,
		 * though.
		 */
		rwb->cur_win_nsec = div_u64(rwb->win_nsec << 4,
					int_sqrt((rqd->scale_step + 1) << 8));
	} else {
		/*
		 * For step < 0, we don't want to increase/decrease the
		 * window size.
		 */
		rwb->cur_win_nsec = rwb->win_nsec;
	}

	blk_stat_activate_nsecs(rwb->cb, rwb->cur_win_nsec);
}

static void wb_timer_fn(struct blk_stat_callback *cb)
{
	struct rq_wb *rwb = cb->data;
	struct rq_depth *rqd = &rwb->rq_depth;
	unsigned int inflight = wbt_inflight(rwb);
	int status;

	if (!rwb->rqos.disk)
		return;

	status = latency_exceeded(rwb, cb->stat);

	trace_wbt_timer(rwb->rqos.disk->bdi, status, rqd->scale_step, inflight);

	/*
	 * If we exceeded the latency target, step down. If we did not,
	 * step one level up. If we don't know enough to say either exceeded
	 * or ok, then don't do anything.
	 */
	switch (status) {
	case LAT_EXCEEDED:
		scale_down(rwb, true);
		break;
	case LAT_OK:
		scale_up(rwb);
		break;
	case LAT_UNKNOWN_WRITES:
		/*
		 * We started a the center step, but don't have a valid
		 * read/write sample, but we do have writes going on.
		 * Allow step to go negative, to increase write perf.
		 */
		scale_up(rwb);
		break;
	case LAT_UNKNOWN:
		if (++rwb->unknown_cnt < RWB_UNKNOWN_BUMP)
			break;
		/*
		 * We get here when previously scaled reduced depth, and we
		 * currently don't have a valid read/write sample. For that
		 * case, slowly return to center state (step == 0).
		 */
		if (rqd->scale_step > 0)
			scale_up(rwb);
		else if (rqd->scale_step < 0)
			scale_down(rwb, false);
		break;
	default:
		break;
	}

	/*
	 * Re-arm timer, if we have IO in flight
	 */
	if (rqd->scale_step || inflight)
		rwb_arm_timer(rwb);
}

static void wbt_update_limits(struct rq_wb *rwb)
{
	struct rq_depth *rqd = &rwb->rq_depth;

	rqd->scale_step = 0;
	rqd->scaled_max = false;

	rq_depth_calc_max_depth(rqd);
	calc_wb_limits(rwb);

	rwb_wake_all(rwb);
}

bool wbt_disabled(struct request_queue *q)
{
	struct rq_qos *rqos = wbt_rq_qos(q);

	return !rqos || !rwb_enabled(RQWB(rqos));
}

u64 wbt_get_min_lat(struct request_queue *q)
{
	struct rq_qos *rqos = wbt_rq_qos(q);
	if (!rqos)
		return 0;
	return RQWB(rqos)->min_lat_nsec;
}

void wbt_set_min_lat(struct request_queue *q, u64 val)
{
	struct rq_qos *rqos = wbt_rq_qos(q);
	if (!rqos)
		return;

	RQWB(rqos)->min_lat_nsec = val;
	if (val)
		RQWB(rqos)->enable_state = WBT_STATE_ON_MANUAL;
	else
		RQWB(rqos)->enable_state = WBT_STATE_OFF_MANUAL;

	wbt_update_limits(RQWB(rqos));
}


static bool close_io(struct rq_wb *rwb)
{
	const unsigned long now = jiffies;

	return time_before(now, rwb->last_issue + HZ / 10) ||
		time_before(now, rwb->last_comp + HZ / 10);
}

#define REQ_HIPRIO	(REQ_SYNC | REQ_META | REQ_PRIO)

static inline unsigned int get_limit(struct rq_wb *rwb, blk_opf_t opf)
{
	unsigned int limit;

	if ((opf & REQ_OP_MASK) == REQ_OP_DISCARD)
		return rwb->wb_background;

	/*
	 * At this point we know it's a buffered write. If this is
	 * kswapd trying to free memory, or REQ_SYNC is set, then
	 * it's WB_SYNC_ALL writeback, and we'll use the max limit for
	 * that. If the write is marked as a background write, then use
	 * the idle limit, or go to normal if we haven't had competing
	 * IO for a bit.
	 */
	if ((opf & REQ_HIPRIO) || wb_recent_wait(rwb) || current_is_kswapd())
		limit = rwb->rq_depth.max_depth;
	else if ((opf & REQ_BACKGROUND) || close_io(rwb)) {
		/*
		 * If less than 100ms since we completed unrelated IO,
		 * limit us to half the depth for background writeback.
		 */
		limit = rwb->wb_background;
	} else
		limit = rwb->wb_normal;

	return limit;
}

struct wbt_wait_data {
	struct rq_wb *rwb;
	enum wbt_flags wb_acct;
	blk_opf_t opf;
};

static bool wbt_inflight_cb(struct rq_wait *rqw, void *private_data)
{
	struct wbt_wait_data *data = private_data;
	return rq_wait_inc_below(rqw, get_limit(data->rwb, data->opf));
}

static void wbt_cleanup_cb(struct rq_wait *rqw, void *private_data)
{
	struct wbt_wait_data *data = private_data;
	wbt_rqw_done(data->rwb, rqw, data->wb_acct);
}

/*
 * Block if we will exceed our limit, or if we are currently waiting for
 * the timer to kick off queuing again.
 */
static void __wbt_wait(struct rq_wb *rwb, enum wbt_flags wb_acct,
		       blk_opf_t opf)
{
	struct rq_wait *rqw = get_rq_wait(rwb, wb_acct);
	struct wbt_wait_data data = {
		.rwb = rwb,
		.wb_acct = wb_acct,
		.opf = opf,
	};

	rq_qos_wait(rqw, &data, wbt_inflight_cb, wbt_cleanup_cb);
}

static inline bool wbt_should_throttle(struct bio *bio)
{
	switch (bio_op(bio)) {
	case REQ_OP_WRITE:
		/*
		 * Don't throttle WRITE_ODIRECT
		 */
		if ((bio->bi_opf & (REQ_SYNC | REQ_IDLE)) ==
		    (REQ_SYNC | REQ_IDLE))
			return false;
		fallthrough;
	case REQ_OP_DISCARD:
		return true;
	default:
		return false;
	}
}

static enum wbt_flags bio_to_wbt_flags(struct rq_wb *rwb, struct bio *bio)
{
	enum wbt_flags flags = 0;

	if (!rwb_enabled(rwb))
		return 0;

	if (bio_op(bio) == REQ_OP_READ) {
		flags = WBT_READ;
	} else if (wbt_should_throttle(bio)) {
		if (current_is_kswapd())
			flags |= WBT_KSWAPD;
		if (bio_op(bio) == REQ_OP_DISCARD)
			flags |= WBT_DISCARD;
		flags |= WBT_TRACKED;
	}
	return flags;
}

static void wbt_cleanup(struct rq_qos *rqos, struct bio *bio)
{
	struct rq_wb *rwb = RQWB(rqos);
	enum wbt_flags flags = bio_to_wbt_flags(rwb, bio);
	__wbt_done(rqos, flags);
}

/*
 * May sleep, if we have exceeded the writeback limits. Caller can pass
 * in an irq held spinlock, if it holds one when calling this function.
 * If we do sleep, we'll release and re-grab it.
 */
static void wbt_wait(struct rq_qos *rqos, struct bio *bio)
{
	struct rq_wb *rwb = RQWB(rqos);
	enum wbt_flags flags;

	flags = bio_to_wbt_flags(rwb, bio);
	if (!(flags & WBT_TRACKED)) {
		if (flags & WBT_READ)
			wb_timestamp(rwb, &rwb->last_issue);
		return;
	}

	__wbt_wait(rwb, flags, bio->bi_opf);

	if (!blk_stat_is_active(rwb->cb))
		rwb_arm_timer(rwb);
}

static void wbt_track(struct rq_qos *rqos, struct request *rq, struct bio *bio)
{
	struct rq_wb *rwb = RQWB(rqos);
	rq->wbt_flags |= bio_to_wbt_flags(rwb, bio);
}

static void wbt_issue(struct rq_qos *rqos, struct request *rq)
{
	struct rq_wb *rwb = RQWB(rqos);

	if (!rwb_enabled(rwb))
		return;

	/*
	 * Track sync issue, in case it takes a long time to complete. Allows us
	 * to react quicker, if a sync IO takes a long time to complete. Note
	 * that this is just a hint. The request can go away when it completes,
	 * so it's important we never dereference it. We only use the address to
	 * compare with, which is why we store the sync_issue time locally.
	 */
	if (wbt_is_read(rq) && !rwb->sync_issue) {
		rwb->sync_cookie = rq;
		rwb->sync_issue = rq->io_start_time_ns;
	}
}

static void wbt_requeue(struct rq_qos *rqos, struct request *rq)
{
	struct rq_wb *rwb = RQWB(rqos);
	if (!rwb_enabled(rwb))
		return;
	if (rq == rwb->sync_cookie) {
		rwb->sync_issue = 0;
		rwb->sync_cookie = NULL;
	}
}

/*
 * Enable wbt if defaults are configured that way
 */
void wbt_enable_default(struct gendisk *disk)
{
	struct request_queue *q = disk->queue;
	struct rq_qos *rqos;
	bool enable = IS_ENABLED(CONFIG_BLK_WBT_MQ);

	if (q->elevator &&
	    test_bit(ELEVATOR_FLAG_DISABLE_WBT, &q->elevator->flags))
		enable = false;

	/* Throttling already enabled? */
	rqos = wbt_rq_qos(q);
	if (rqos) {
		if (enable && RQWB(rqos)->enable_state == WBT_STATE_OFF_DEFAULT)
			RQWB(rqos)->enable_state = WBT_STATE_ON_DEFAULT;
		return;
	}

	/* Queue not registered? Maybe shutting down... */
	if (!blk_queue_registered(q))
		return;

	if (queue_is_mq(q) && enable)
		wbt_init(disk);
}
EXPORT_SYMBOL_GPL(wbt_enable_default);

u64 wbt_default_latency_nsec(struct request_queue *q)
{
	/*
	 * We default to 2msec for non-rotational storage, and 75msec
	 * for rotational storage.
	 */
	if (blk_queue_nonrot(q))
		return 2000000ULL;
	else
		return 75000000ULL;
}

static int wbt_data_dir(const struct request *rq)
{
	const enum req_op op = req_op(rq);

	if (op == REQ_OP_READ)
		return READ;
	else if (op_is_write(op))
		return WRITE;

	/* don't account */
	return -1;
}

static void wbt_queue_depth_changed(struct rq_qos *rqos)
{
	RQWB(rqos)->rq_depth.queue_depth = blk_queue_depth(rqos->disk->queue);
	wbt_update_limits(RQWB(rqos));
}

static void wbt_exit(struct rq_qos *rqos)
{
	struct rq_wb *rwb = RQWB(rqos);

	blk_stat_remove_callback(rqos->disk->queue, rwb->cb);
	blk_stat_free_callback(rwb->cb);
	kfree(rwb);
}

/*
 * Disable wbt, if enabled by default.
 */
void wbt_disable_default(struct gendisk *disk)
{
	struct rq_qos *rqos = wbt_rq_qos(disk->queue);
	struct rq_wb *rwb;
	if (!rqos)
		return;
	rwb = RQWB(rqos);
	if (rwb->enable_state == WBT_STATE_ON_DEFAULT) {
		blk_stat_deactivate(rwb->cb);
		rwb->enable_state = WBT_STATE_OFF_DEFAULT;
	}
}
EXPORT_SYMBOL_GPL(wbt_disable_default);

#ifdef CONFIG_BLK_DEBUG_FS
static int wbt_curr_win_nsec_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);

	seq_printf(m, "%llu\n", rwb->cur_win_nsec);
	return 0;
}

static int wbt_enabled_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);

	seq_printf(m, "%d\n", rwb->enable_state);
	return 0;
}

static int wbt_id_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;

	seq_printf(m, "%u\n", rqos->id);
	return 0;
}

static int wbt_inflight_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);
	int i;

	for (i = 0; i < WBT_NUM_RWQ; i++)
		seq_printf(m, "%d: inflight %d\n", i,
			   atomic_read(&rwb->rq_wait[i].inflight));
	return 0;
}

static int wbt_min_lat_nsec_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);

	seq_printf(m, "%lu\n", rwb->min_lat_nsec);
	return 0;
}

static int wbt_unknown_cnt_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);

	seq_printf(m, "%u\n", rwb->unknown_cnt);
	return 0;
}

static int wbt_normal_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);

	seq_printf(m, "%u\n", rwb->wb_normal);
	return 0;
}

static int wbt_background_show(void *data, struct seq_file *m)
{
	struct rq_qos *rqos = data;
	struct rq_wb *rwb = RQWB(rqos);

	seq_printf(m, "%u\n", rwb->wb_background);
	return 0;
}

static const struct blk_mq_debugfs_attr wbt_debugfs_attrs[] = {
	{"curr_win_nsec", 0400, wbt_curr_win_nsec_show},
	{"enabled", 0400, wbt_enabled_show},
	{"id", 0400, wbt_id_show},
	{"inflight", 0400, wbt_inflight_show},
	{"min_lat_nsec", 0400, wbt_min_lat_nsec_show},
	{"unknown_cnt", 0400, wbt_unknown_cnt_show},
	{"wb_normal", 0400, wbt_normal_show},
	{"wb_background", 0400, wbt_background_show},
	{},
};
#endif

static const struct rq_qos_ops wbt_rqos_ops = {
	.throttle = wbt_wait,
	.issue = wbt_issue,
	.track = wbt_track,
	.requeue = wbt_requeue,
	.done = wbt_done,
	.cleanup = wbt_cleanup,
	.queue_depth_changed = wbt_queue_depth_changed,
	.exit = wbt_exit,
#ifdef CONFIG_BLK_DEBUG_FS
	.debugfs_attrs = wbt_debugfs_attrs,
#endif
};

int wbt_init(struct gendisk *disk)
{
	struct request_queue *q = disk->queue;
	struct rq_wb *rwb;
	int i;
	int ret;

	rwb = kzalloc(sizeof(*rwb), GFP_KERNEL);
	if (!rwb)
		return -ENOMEM;

	rwb->cb = blk_stat_alloc_callback(wb_timer_fn, wbt_data_dir, 2, rwb);
	if (!rwb->cb) {
		kfree(rwb);
		return -ENOMEM;
	}

	for (i = 0; i < WBT_NUM_RWQ; i++)
		rq_wait_init(&rwb->rq_wait[i]);

	rwb->last_comp = rwb->last_issue = jiffies;
	rwb->win_nsec = RWB_WINDOW_NSEC;
	rwb->enable_state = WBT_STATE_ON_DEFAULT;
	rwb->rq_depth.default_depth = RWB_DEF_DEPTH;
	rwb->min_lat_nsec = wbt_default_latency_nsec(q);
	rwb->rq_depth.queue_depth = blk_queue_depth(q);
	wbt_update_limits(rwb);

	/*
	 * Assign rwb and add the stats callback.
	 */
	mutex_lock(&q->rq_qos_mutex);
	ret = rq_qos_add(&rwb->rqos, disk, RQ_QOS_WBT, &wbt_rqos_ops);
	mutex_unlock(&q->rq_qos_mutex);
	if (ret)
		goto err_free;

	blk_stat_add_callback(q, rwb->cb);

	return 0;

err_free:
	blk_stat_free_callback(rwb->cb);
	kfree(rwb);
	return ret;

}