Training courses

Kernel and Embedded Linux

Bootlin training courses

Embedded Linux, kernel,
Yocto Project, Buildroot, real-time,
graphics, boot time, debugging...

Bootlin logo

Elixir Cross Referencer

  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
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright 2009 Sun Microsystems, Inc.  All rights reserved.
 * Use is subject to license terms.
 */

/*
 * Copyright (c) 2012, 2014 by Delphix. All rights reserved.
 * Copyright (c) 2014 Integros [integros.com]
 */

#include <sys/zfs_context.h>
#include <sys/vdev_impl.h>
#include <sys/spa_impl.h>
#include <sys/zio.h>
#include <sys/avl.h>
#include <sys/dsl_pool.h>
#include <sys/metaslab_impl.h>

/*
 * ZFS I/O Scheduler
 * ---------------
 *
 * ZFS issues I/O operations to leaf vdevs to satisfy and complete zios.  The
 * I/O scheduler determines when and in what order those operations are
 * issued.  The I/O scheduler divides operations into six I/O classes
 * prioritized in the following order: sync read, sync write, async read,
 * async write, scrub/resilver and trim.  Each queue defines the minimum and
 * maximum number of concurrent operations that may be issued to the device.
 * In addition, the device has an aggregate maximum. Note that the sum of the
 * per-queue minimums must not exceed the aggregate maximum, and if the
 * aggregate maximum is equal to or greater than the sum of the per-queue
 * maximums, the per-queue minimum has no effect.
 *
 * For many physical devices, throughput increases with the number of
 * concurrent operations, but latency typically suffers. Further, physical
 * devices typically have a limit at which more concurrent operations have no
 * effect on throughput or can actually cause it to decrease.
 *
 * The scheduler selects the next operation to issue by first looking for an
 * I/O class whose minimum has not been satisfied. Once all are satisfied and
 * the aggregate maximum has not been hit, the scheduler looks for classes
 * whose maximum has not been satisfied. Iteration through the I/O classes is
 * done in the order specified above. No further operations are issued if the
 * aggregate maximum number of concurrent operations has been hit or if there
 * are no operations queued for an I/O class that has not hit its maximum.
 * Every time an I/O is queued or an operation completes, the I/O scheduler
 * looks for new operations to issue.
 *
 * All I/O classes have a fixed maximum number of outstanding operations
 * except for the async write class. Asynchronous writes represent the data
 * that is committed to stable storage during the syncing stage for
 * transaction groups (see txg.c). Transaction groups enter the syncing state
 * periodically so the number of queued async writes will quickly burst up and
 * then bleed down to zero. Rather than servicing them as quickly as possible,
 * the I/O scheduler changes the maximum number of active async write I/Os
 * according to the amount of dirty data in the pool (see dsl_pool.c). Since
 * both throughput and latency typically increase with the number of
 * concurrent operations issued to physical devices, reducing the burstiness
 * in the number of concurrent operations also stabilizes the response time of
 * operations from other -- and in particular synchronous -- queues. In broad
 * strokes, the I/O scheduler will issue more concurrent operations from the
 * async write queue as there's more dirty data in the pool.
 *
 * Async Writes
 *
 * The number of concurrent operations issued for the async write I/O class
 * follows a piece-wise linear function defined by a few adjustable points.
 *
 *        |                   o---------| <-- zfs_vdev_async_write_max_active
 *   ^    |                  /^         |
 *   |    |                 / |         |
 * active |                /  |         |
 *  I/O   |               /   |         |
 * count  |              /    |         |
 *        |             /     |         |
 *        |------------o      |         | <-- zfs_vdev_async_write_min_active
 *       0|____________^______|_________|
 *        0%           |      |       100% of zfs_dirty_data_max
 *                     |      |
 *                     |      `-- zfs_vdev_async_write_active_max_dirty_percent
 *                     `--------- zfs_vdev_async_write_active_min_dirty_percent
 *
 * Until the amount of dirty data exceeds a minimum percentage of the dirty
 * data allowed in the pool, the I/O scheduler will limit the number of
 * concurrent operations to the minimum. As that threshold is crossed, the
 * number of concurrent operations issued increases linearly to the maximum at
 * the specified maximum percentage of the dirty data allowed in the pool.
 *
 * Ideally, the amount of dirty data on a busy pool will stay in the sloped
 * part of the function between zfs_vdev_async_write_active_min_dirty_percent
 * and zfs_vdev_async_write_active_max_dirty_percent. If it exceeds the
 * maximum percentage, this indicates that the rate of incoming data is
 * greater than the rate that the backend storage can handle. In this case, we
 * must further throttle incoming writes (see dmu_tx_delay() for details).
 */

/*
 * The maximum number of I/Os active to each device.  Ideally, this will be >=
 * the sum of each queue's max_active.  It must be at least the sum of each
 * queue's min_active.
 */
uint32_t zfs_vdev_max_active = 1000;

/*
 * Per-queue limits on the number of I/Os active to each device.  If the
 * sum of the queue's max_active is < zfs_vdev_max_active, then the
 * min_active comes into play.  We will send min_active from each queue,
 * and then select from queues in the order defined by zio_priority_t.
 *
 * In general, smaller max_active's will lead to lower latency of synchronous
 * operations.  Larger max_active's may lead to higher overall throughput,
 * depending on underlying storage.
 *
 * The ratio of the queues' max_actives determines the balance of performance
 * between reads, writes, and scrubs.  E.g., increasing
 * zfs_vdev_scrub_max_active will cause the scrub or resilver to complete
 * more quickly, but reads and writes to have higher latency and lower
 * throughput.
 */
uint32_t zfs_vdev_sync_read_min_active = 10;
uint32_t zfs_vdev_sync_read_max_active = 10;
uint32_t zfs_vdev_sync_write_min_active = 10;
uint32_t zfs_vdev_sync_write_max_active = 10;
uint32_t zfs_vdev_async_read_min_active = 1;
uint32_t zfs_vdev_async_read_max_active = 3;
uint32_t zfs_vdev_async_write_min_active = 1;
uint32_t zfs_vdev_async_write_max_active = 10;
uint32_t zfs_vdev_scrub_min_active = 1;
uint32_t zfs_vdev_scrub_max_active = 2;
uint32_t zfs_vdev_trim_min_active = 1;
/*
 * TRIM max active is large in comparison to the other values due to the fact
 * that TRIM IOs are coalesced at the device layer. This value is set such
 * that a typical SSD can process the queued IOs in a single request.
 */
uint32_t zfs_vdev_trim_max_active = 64;


/*
 * When the pool has less than zfs_vdev_async_write_active_min_dirty_percent
 * dirty data, use zfs_vdev_async_write_min_active.  When it has more than
 * zfs_vdev_async_write_active_max_dirty_percent, use
 * zfs_vdev_async_write_max_active. The value is linearly interpolated
 * between min and max.
 */
int zfs_vdev_async_write_active_min_dirty_percent = 30;
int zfs_vdev_async_write_active_max_dirty_percent = 60;

/*
 * To reduce IOPs, we aggregate small adjacent I/Os into one large I/O.
 * For read I/Os, we also aggregate across small adjacency gaps; for writes
 * we include spans of optional I/Os to aid aggregation at the disk even when
 * they aren't able to help us aggregate at this level.
 */
int zfs_vdev_aggregation_limit = SPA_OLD_MAXBLOCKSIZE;
int zfs_vdev_read_gap_limit = 32 << 10;
int zfs_vdev_write_gap_limit = 4 << 10;

/*
 * Define the queue depth percentage for each top-level. This percentage is
 * used in conjunction with zfs_vdev_async_max_active to determine how many
 * allocations a specific top-level vdev should handle. Once the queue depth
 * reaches zfs_vdev_queue_depth_pct * zfs_vdev_async_write_max_active / 100
 * then allocator will stop allocating blocks on that top-level device.
 * The default kernel setting is 1000% which will yield 100 allocations per
 * device. For userland testing, the default setting is 300% which equates
 * to 30 allocations per device.
 */
#ifdef _KERNEL
int zfs_vdev_queue_depth_pct = 1000;
#else
int zfs_vdev_queue_depth_pct = 300;
#endif


#ifdef __FreeBSD__
#ifdef _KERNEL
SYSCTL_DECL(_vfs_zfs_vdev);

static int sysctl_zfs_async_write_active_min_dirty_percent(SYSCTL_HANDLER_ARGS);
SYSCTL_PROC(_vfs_zfs_vdev, OID_AUTO, async_write_active_min_dirty_percent,
    CTLTYPE_UINT | CTLFLAG_MPSAFE | CTLFLAG_RWTUN, 0, sizeof(int),
    sysctl_zfs_async_write_active_min_dirty_percent, "I",
    "Percentage of async write dirty data below which "
    "async_write_min_active is used.");

static int sysctl_zfs_async_write_active_max_dirty_percent(SYSCTL_HANDLER_ARGS);
SYSCTL_PROC(_vfs_zfs_vdev, OID_AUTO, async_write_active_max_dirty_percent,
    CTLTYPE_UINT | CTLFLAG_MPSAFE | CTLFLAG_RWTUN, 0, sizeof(int),
    sysctl_zfs_async_write_active_max_dirty_percent, "I",
    "Percentage of async write dirty data above which "
    "async_write_max_active is used.");

SYSCTL_UINT(_vfs_zfs_vdev, OID_AUTO, max_active, CTLFLAG_RWTUN,
    &zfs_vdev_max_active, 0,
    "The maximum number of I/Os of all types active for each device.");

#define ZFS_VDEV_QUEUE_KNOB_MIN(name)					\
SYSCTL_UINT(_vfs_zfs_vdev, OID_AUTO, name ## _min_active, CTLFLAG_RWTUN,\
    &zfs_vdev_ ## name ## _min_active, 0,				\
    "Initial number of I/O requests of type " #name			\
    " active for each device");

#define ZFS_VDEV_QUEUE_KNOB_MAX(name)					\
SYSCTL_UINT(_vfs_zfs_vdev, OID_AUTO, name ## _max_active, CTLFLAG_RWTUN,\
    &zfs_vdev_ ## name ## _max_active, 0,				\
    "Maximum number of I/O requests of type " #name			\
    " active for each device");

ZFS_VDEV_QUEUE_KNOB_MIN(sync_read);
ZFS_VDEV_QUEUE_KNOB_MAX(sync_read);
ZFS_VDEV_QUEUE_KNOB_MIN(sync_write);
ZFS_VDEV_QUEUE_KNOB_MAX(sync_write);
ZFS_VDEV_QUEUE_KNOB_MIN(async_read);
ZFS_VDEV_QUEUE_KNOB_MAX(async_read);
ZFS_VDEV_QUEUE_KNOB_MIN(async_write);
ZFS_VDEV_QUEUE_KNOB_MAX(async_write);
ZFS_VDEV_QUEUE_KNOB_MIN(scrub);
ZFS_VDEV_QUEUE_KNOB_MAX(scrub);
ZFS_VDEV_QUEUE_KNOB_MIN(trim);
ZFS_VDEV_QUEUE_KNOB_MAX(trim);

#undef ZFS_VDEV_QUEUE_KNOB

SYSCTL_INT(_vfs_zfs_vdev, OID_AUTO, aggregation_limit, CTLFLAG_RWTUN,
    &zfs_vdev_aggregation_limit, 0,
    "I/O requests are aggregated up to this size");
SYSCTL_INT(_vfs_zfs_vdev, OID_AUTO, read_gap_limit, CTLFLAG_RWTUN,
    &zfs_vdev_read_gap_limit, 0,
    "Acceptable gap between two reads being aggregated");
SYSCTL_INT(_vfs_zfs_vdev, OID_AUTO, write_gap_limit, CTLFLAG_RWTUN,
    &zfs_vdev_write_gap_limit, 0,
    "Acceptable gap between two writes being aggregated");
SYSCTL_INT(_vfs_zfs_vdev, OID_AUTO, queue_depth_pct, CTLFLAG_RWTUN,
    &zfs_vdev_queue_depth_pct, 0,
    "Queue depth percentage for each top-level");

static int
sysctl_zfs_async_write_active_min_dirty_percent(SYSCTL_HANDLER_ARGS)
{
	int val, err;

	val = zfs_vdev_async_write_active_min_dirty_percent;
	err = sysctl_handle_int(oidp, &val, 0, req);
	if (err != 0 || req->newptr == NULL)
		return (err);
	
	if (val < 0 || val > 100 ||
	    val >= zfs_vdev_async_write_active_max_dirty_percent)
		return (EINVAL);

	zfs_vdev_async_write_active_min_dirty_percent = val;

	return (0);
}

static int
sysctl_zfs_async_write_active_max_dirty_percent(SYSCTL_HANDLER_ARGS)
{
	int val, err;

	val = zfs_vdev_async_write_active_max_dirty_percent;
	err = sysctl_handle_int(oidp, &val, 0, req);
	if (err != 0 || req->newptr == NULL)
		return (err);

	if (val < 0 || val > 100 ||
	    val <= zfs_vdev_async_write_active_min_dirty_percent)
		return (EINVAL);

	zfs_vdev_async_write_active_max_dirty_percent = val;

	return (0);
}
#endif
#endif

int
vdev_queue_offset_compare(const void *x1, const void *x2)
{
	const zio_t *z1 = x1;
	const zio_t *z2 = x2;

	if (z1->io_offset < z2->io_offset)
		return (-1);
	if (z1->io_offset > z2->io_offset)
		return (1);

	if (z1 < z2)
		return (-1);
	if (z1 > z2)
		return (1);

	return (0);
}

static inline avl_tree_t *
vdev_queue_class_tree(vdev_queue_t *vq, zio_priority_t p)
{
	return (&vq->vq_class[p].vqc_queued_tree);
}

static inline avl_tree_t *
vdev_queue_type_tree(vdev_queue_t *vq, zio_type_t t)
{
	if (t == ZIO_TYPE_READ)
		return (&vq->vq_read_offset_tree);
	else if (t == ZIO_TYPE_WRITE)
		return (&vq->vq_write_offset_tree);
	else
		return (NULL);
}

int
vdev_queue_timestamp_compare(const void *x1, const void *x2)
{
	const zio_t *z1 = x1;
	const zio_t *z2 = x2;

	if (z1->io_timestamp < z2->io_timestamp)
		return (-1);
	if (z1->io_timestamp > z2->io_timestamp)
		return (1);

	if (z1->io_offset < z2->io_offset)
		return (-1);
	if (z1->io_offset > z2->io_offset)
		return (1);

	if (z1 < z2)
		return (-1);
	if (z1 > z2)
		return (1);

	return (0);
}

void
vdev_queue_init(vdev_t *vd)
{
	vdev_queue_t *vq = &vd->vdev_queue;

	mutex_init(&vq->vq_lock, NULL, MUTEX_DEFAULT, NULL);
	vq->vq_vdev = vd;

	avl_create(&vq->vq_active_tree, vdev_queue_offset_compare,
	    sizeof (zio_t), offsetof(struct zio, io_queue_node));
	avl_create(vdev_queue_type_tree(vq, ZIO_TYPE_READ),
	    vdev_queue_offset_compare, sizeof (zio_t),
	    offsetof(struct zio, io_offset_node));
	avl_create(vdev_queue_type_tree(vq, ZIO_TYPE_WRITE),
	    vdev_queue_offset_compare, sizeof (zio_t),
	    offsetof(struct zio, io_offset_node));

	for (zio_priority_t p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) {
		int (*compfn) (const void *, const void *);

		/*
		 * The synchronous i/o queues are dispatched in FIFO rather
		 * than LBA order.  This provides more consistent latency for
		 * these i/os.
		 */
		if (p == ZIO_PRIORITY_SYNC_READ || p == ZIO_PRIORITY_SYNC_WRITE)
			compfn = vdev_queue_timestamp_compare;
		else
			compfn = vdev_queue_offset_compare;

		avl_create(vdev_queue_class_tree(vq, p), compfn,
		    sizeof (zio_t), offsetof(struct zio, io_queue_node));
	}

	vq->vq_lastoffset = 0;
}

void
vdev_queue_fini(vdev_t *vd)
{
	vdev_queue_t *vq = &vd->vdev_queue;

	for (zio_priority_t p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++)
		avl_destroy(vdev_queue_class_tree(vq, p));
	avl_destroy(&vq->vq_active_tree);
	avl_destroy(vdev_queue_type_tree(vq, ZIO_TYPE_READ));
	avl_destroy(vdev_queue_type_tree(vq, ZIO_TYPE_WRITE));

	mutex_destroy(&vq->vq_lock);
}

static void
vdev_queue_io_add(vdev_queue_t *vq, zio_t *zio)
{
	spa_t *spa = zio->io_spa;
	avl_tree_t *qtt;

	ASSERT(MUTEX_HELD(&vq->vq_lock));
	ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE);
	avl_add(vdev_queue_class_tree(vq, zio->io_priority), zio);
	qtt = vdev_queue_type_tree(vq, zio->io_type);
	if (qtt)
		avl_add(qtt, zio);

#ifdef illumos
	mutex_enter(&spa->spa_iokstat_lock);
	spa->spa_queue_stats[zio->io_priority].spa_queued++;
	if (spa->spa_iokstat != NULL)
		kstat_waitq_enter(spa->spa_iokstat->ks_data);
	mutex_exit(&spa->spa_iokstat_lock);
#endif
}

static void
vdev_queue_io_remove(vdev_queue_t *vq, zio_t *zio)
{
	spa_t *spa = zio->io_spa;
	avl_tree_t *qtt;

	ASSERT(MUTEX_HELD(&vq->vq_lock));
	ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE);
	avl_remove(vdev_queue_class_tree(vq, zio->io_priority), zio);
	qtt = vdev_queue_type_tree(vq, zio->io_type);
	if (qtt)
		avl_remove(qtt, zio);

#ifdef illumos
	mutex_enter(&spa->spa_iokstat_lock);
	ASSERT3U(spa->spa_queue_stats[zio->io_priority].spa_queued, >, 0);
	spa->spa_queue_stats[zio->io_priority].spa_queued--;
	if (spa->spa_iokstat != NULL)
		kstat_waitq_exit(spa->spa_iokstat->ks_data);
	mutex_exit(&spa->spa_iokstat_lock);
#endif
}

static void
vdev_queue_pending_add(vdev_queue_t *vq, zio_t *zio)
{
	spa_t *spa = zio->io_spa;
	ASSERT(MUTEX_HELD(&vq->vq_lock));
	ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE);
	vq->vq_class[zio->io_priority].vqc_active++;
	avl_add(&vq->vq_active_tree, zio);

#ifdef illumos
	mutex_enter(&spa->spa_iokstat_lock);
	spa->spa_queue_stats[zio->io_priority].spa_active++;
	if (spa->spa_iokstat != NULL)
		kstat_runq_enter(spa->spa_iokstat->ks_data);
	mutex_exit(&spa->spa_iokstat_lock);
#endif
}

static void
vdev_queue_pending_remove(vdev_queue_t *vq, zio_t *zio)
{
	spa_t *spa = zio->io_spa;
	ASSERT(MUTEX_HELD(&vq->vq_lock));
	ASSERT3U(zio->io_priority, <, ZIO_PRIORITY_NUM_QUEUEABLE);
	vq->vq_class[zio->io_priority].vqc_active--;
	avl_remove(&vq->vq_active_tree, zio);

#ifdef illumos
	mutex_enter(&spa->spa_iokstat_lock);
	ASSERT3U(spa->spa_queue_stats[zio->io_priority].spa_active, >, 0);
	spa->spa_queue_stats[zio->io_priority].spa_active--;
	if (spa->spa_iokstat != NULL) {
		kstat_io_t *ksio = spa->spa_iokstat->ks_data;

		kstat_runq_exit(spa->spa_iokstat->ks_data);
		if (zio->io_type == ZIO_TYPE_READ) {
			ksio->reads++;
			ksio->nread += zio->io_size;
		} else if (zio->io_type == ZIO_TYPE_WRITE) {
			ksio->writes++;
			ksio->nwritten += zio->io_size;
		}
	}
	mutex_exit(&spa->spa_iokstat_lock);
#endif
}

static void
vdev_queue_agg_io_done(zio_t *aio)
{
	if (aio->io_type == ZIO_TYPE_READ) {
		zio_t *pio;
		zio_link_t *zl = NULL;
		while ((pio = zio_walk_parents(aio, &zl)) != NULL) {
			bcopy((char *)aio->io_data + (pio->io_offset -
			    aio->io_offset), pio->io_data, pio->io_size);
		}
	}

	zio_buf_free(aio->io_data, aio->io_size);
}

static int
vdev_queue_class_min_active(zio_priority_t p)
{
	switch (p) {
	case ZIO_PRIORITY_SYNC_READ:
		return (zfs_vdev_sync_read_min_active);
	case ZIO_PRIORITY_SYNC_WRITE:
		return (zfs_vdev_sync_write_min_active);
	case ZIO_PRIORITY_ASYNC_READ:
		return (zfs_vdev_async_read_min_active);
	case ZIO_PRIORITY_ASYNC_WRITE:
		return (zfs_vdev_async_write_min_active);
	case ZIO_PRIORITY_SCRUB:
		return (zfs_vdev_scrub_min_active);
	case ZIO_PRIORITY_TRIM:
		return (zfs_vdev_trim_min_active);
	default:
		panic("invalid priority %u", p);
		return (0);
	}
}

static __noinline int
vdev_queue_max_async_writes(spa_t *spa)
{
	int writes;
	uint64_t dirty = spa->spa_dsl_pool->dp_dirty_total;
	uint64_t min_bytes = zfs_dirty_data_max *
	    zfs_vdev_async_write_active_min_dirty_percent / 100;
	uint64_t max_bytes = zfs_dirty_data_max *
	    zfs_vdev_async_write_active_max_dirty_percent / 100;

	/*
	 * Sync tasks correspond to interactive user actions. To reduce the
	 * execution time of those actions we push data out as fast as possible.
	 */
	if (spa_has_pending_synctask(spa)) {
		return (zfs_vdev_async_write_max_active);
	}

	if (dirty < min_bytes)
		return (zfs_vdev_async_write_min_active);
	if (dirty > max_bytes)
		return (zfs_vdev_async_write_max_active);

	/*
	 * linear interpolation:
	 * slope = (max_writes - min_writes) / (max_bytes - min_bytes)
	 * move right by min_bytes
	 * move up by min_writes
	 */
	writes = (dirty - min_bytes) *
	    (zfs_vdev_async_write_max_active -
	    zfs_vdev_async_write_min_active) /
	    (max_bytes - min_bytes) +
	    zfs_vdev_async_write_min_active;
	ASSERT3U(writes, >=, zfs_vdev_async_write_min_active);
	ASSERT3U(writes, <=, zfs_vdev_async_write_max_active);
	return (writes);
}

static int
vdev_queue_class_max_active(spa_t *spa, zio_priority_t p)
{
	switch (p) {
	case ZIO_PRIORITY_SYNC_READ:
		return (zfs_vdev_sync_read_max_active);
	case ZIO_PRIORITY_SYNC_WRITE:
		return (zfs_vdev_sync_write_max_active);
	case ZIO_PRIORITY_ASYNC_READ:
		return (zfs_vdev_async_read_max_active);
	case ZIO_PRIORITY_ASYNC_WRITE:
		return (vdev_queue_max_async_writes(spa));
	case ZIO_PRIORITY_SCRUB:
		return (zfs_vdev_scrub_max_active);
	case ZIO_PRIORITY_TRIM:
		return (zfs_vdev_trim_max_active);
	default:
		panic("invalid priority %u", p);
		return (0);
	}
}

/*
 * Return the i/o class to issue from, or ZIO_PRIORITY_MAX_QUEUEABLE if
 * there is no eligible class.
 */
static zio_priority_t
vdev_queue_class_to_issue(vdev_queue_t *vq)
{
	spa_t *spa = vq->vq_vdev->vdev_spa;
	zio_priority_t p;

	ASSERT(MUTEX_HELD(&vq->vq_lock));

	if (avl_numnodes(&vq->vq_active_tree) >= zfs_vdev_max_active)
		return (ZIO_PRIORITY_NUM_QUEUEABLE);

	/* find a queue that has not reached its minimum # outstanding i/os */
	for (p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) {
		if (avl_numnodes(vdev_queue_class_tree(vq, p)) > 0 &&
		    vq->vq_class[p].vqc_active <
		    vdev_queue_class_min_active(p))
			return (p);
	}

	/*
	 * If we haven't found a queue, look for one that hasn't reached its
	 * maximum # outstanding i/os.
	 */
	for (p = 0; p < ZIO_PRIORITY_NUM_QUEUEABLE; p++) {
		if (avl_numnodes(vdev_queue_class_tree(vq, p)) > 0 &&
		    vq->vq_class[p].vqc_active <
		    vdev_queue_class_max_active(spa, p))
			return (p);
	}

	/* No eligible queued i/os */
	return (ZIO_PRIORITY_NUM_QUEUEABLE);
}

/*
 * Compute the range spanned by two i/os, which is the endpoint of the last
 * (lio->io_offset + lio->io_size) minus start of the first (fio->io_offset).
 * Conveniently, the gap between fio and lio is given by -IO_SPAN(lio, fio);
 * thus fio and lio are adjacent if and only if IO_SPAN(lio, fio) == 0.
 */
#define	IO_SPAN(fio, lio) ((lio)->io_offset + (lio)->io_size - (fio)->io_offset)
#define	IO_GAP(fio, lio) (-IO_SPAN(lio, fio))

static zio_t *
vdev_queue_aggregate(vdev_queue_t *vq, zio_t *zio)
{
	zio_t *first, *last, *aio, *dio, *mandatory, *nio;
	void *abuf;
	uint64_t maxgap = 0;
	uint64_t size;
	boolean_t stretch;
	avl_tree_t *t;
	enum zio_flag flags;

	ASSERT(MUTEX_HELD(&vq->vq_lock));

	if (zio->io_flags & ZIO_FLAG_DONT_AGGREGATE)
		return (NULL);

	first = last = zio;

	if (zio->io_type == ZIO_TYPE_READ)
		maxgap = zfs_vdev_read_gap_limit;

	/*
	 * We can aggregate I/Os that are sufficiently adjacent and of
	 * the same flavor, as expressed by the AGG_INHERIT flags.
	 * The latter requirement is necessary so that certain
	 * attributes of the I/O, such as whether it's a normal I/O
	 * or a scrub/resilver, can be preserved in the aggregate.
	 * We can include optional I/Os, but don't allow them
	 * to begin a range as they add no benefit in that situation.
	 */

	/*
	 * We keep track of the last non-optional I/O.
	 */
	mandatory = (first->io_flags & ZIO_FLAG_OPTIONAL) ? NULL : first;

	/*
	 * Walk backwards through sufficiently contiguous I/Os
	 * recording the last non-option I/O.
	 */
	flags = zio->io_flags & ZIO_FLAG_AGG_INHERIT;
	t = vdev_queue_type_tree(vq, zio->io_type);
	while (t != NULL && (dio = AVL_PREV(t, first)) != NULL &&
	    (dio->io_flags & ZIO_FLAG_AGG_INHERIT) == flags &&
	    IO_SPAN(dio, last) <= zfs_vdev_aggregation_limit &&
	    IO_GAP(dio, first) <= maxgap) {
		first = dio;
		if (mandatory == NULL && !(first->io_flags & ZIO_FLAG_OPTIONAL))
			mandatory = first;
	}

	/*
	 * Skip any initial optional I/Os.
	 */
	while ((first->io_flags & ZIO_FLAG_OPTIONAL) && first != last) {
		first = AVL_NEXT(t, first);
		ASSERT(first != NULL);
	}

	/*
	 * Walk forward through sufficiently contiguous I/Os.
	 */
	while ((dio = AVL_NEXT(t, last)) != NULL &&
	    (dio->io_flags & ZIO_FLAG_AGG_INHERIT) == flags &&
	    IO_SPAN(first, dio) <= zfs_vdev_aggregation_limit &&
	    IO_GAP(last, dio) <= maxgap) {
		last = dio;
		if (!(last->io_flags & ZIO_FLAG_OPTIONAL))
			mandatory = last;
	}

	/*
	 * Now that we've established the range of the I/O aggregation
	 * we must decide what to do with trailing optional I/Os.
	 * For reads, there's nothing to do. While we are unable to
	 * aggregate further, it's possible that a trailing optional
	 * I/O would allow the underlying device to aggregate with
	 * subsequent I/Os. We must therefore determine if the next
	 * non-optional I/O is close enough to make aggregation
	 * worthwhile.
	 */
	stretch = B_FALSE;
	if (zio->io_type == ZIO_TYPE_WRITE && mandatory != NULL) {
		zio_t *nio = last;
		while ((dio = AVL_NEXT(t, nio)) != NULL &&
		    IO_GAP(nio, dio) == 0 &&
		    IO_GAP(mandatory, dio) <= zfs_vdev_write_gap_limit) {
			nio = dio;
			if (!(nio->io_flags & ZIO_FLAG_OPTIONAL)) {
				stretch = B_TRUE;
				break;
			}
		}
	}

	if (stretch) {
		/* This may be a no-op. */
		dio = AVL_NEXT(t, last);
		dio->io_flags &= ~ZIO_FLAG_OPTIONAL;
	} else {
		while (last != mandatory && last != first) {
			ASSERT(last->io_flags & ZIO_FLAG_OPTIONAL);
			last = AVL_PREV(t, last);
			ASSERT(last != NULL);
		}
	}

	if (first == last)
		return (NULL);

	size = IO_SPAN(first, last);
	ASSERT3U(size, <=, zfs_vdev_aggregation_limit);

	abuf = zio_buf_alloc_nowait(size);
	if (abuf == NULL)
		return (NULL);

	aio = zio_vdev_delegated_io(first->io_vd, first->io_offset,
	    abuf, size, first->io_type, zio->io_priority,
	    flags | ZIO_FLAG_DONT_CACHE | ZIO_FLAG_DONT_QUEUE,
	    vdev_queue_agg_io_done, NULL);
	aio->io_timestamp = first->io_timestamp;

	nio = first;
	do {
		dio = nio;
		nio = AVL_NEXT(t, dio);
		ASSERT3U(dio->io_type, ==, aio->io_type);

		if (dio->io_flags & ZIO_FLAG_NODATA) {
			ASSERT3U(dio->io_type, ==, ZIO_TYPE_WRITE);
			bzero((char *)aio->io_data + (dio->io_offset -
			    aio->io_offset), dio->io_size);
		} else if (dio->io_type == ZIO_TYPE_WRITE) {
			bcopy(dio->io_data, (char *)aio->io_data +
			    (dio->io_offset - aio->io_offset),
			    dio->io_size);
		}

		zio_add_child(dio, aio);
		vdev_queue_io_remove(vq, dio);
		zio_vdev_io_bypass(dio);
		zio_execute(dio);
	} while (dio != last);

	return (aio);
}

static zio_t *
vdev_queue_io_to_issue(vdev_queue_t *vq)
{
	zio_t *zio, *aio;
	zio_priority_t p;
	avl_index_t idx;
	avl_tree_t *tree;
	zio_t *search;

again:
	ASSERT(MUTEX_HELD(&vq->vq_lock));

	p = vdev_queue_class_to_issue(vq);

	if (p == ZIO_PRIORITY_NUM_QUEUEABLE) {
		/* No eligible queued i/os */
		return (NULL);
	}

	/*
	 * For LBA-ordered queues (async / scrub), issue the i/o which follows
	 * the most recently issued i/o in LBA (offset) order.
	 *
	 * For FIFO queues (sync), issue the i/o with the lowest timestamp.
	 */
	tree = vdev_queue_class_tree(vq, p);
	search = kmem_zalloc(sizeof (*search), KM_NOSLEEP);
	if (search) {
		search->io_offset = vq->vq_last_offset + 1;
		VERIFY3P(avl_find(tree, search, &idx), ==, NULL);
		kmem_free(search, sizeof (*search));
		zio = avl_nearest(tree, idx, AVL_AFTER);
	} else {
		/* Can't find nearest, fallback to first */
		zio = NULL;
	}
	if (zio == NULL)
		zio = avl_first(tree);
	ASSERT3U(zio->io_priority, ==, p);

	aio = vdev_queue_aggregate(vq, zio);
	if (aio != NULL)
		zio = aio;
	else
		vdev_queue_io_remove(vq, zio);

	/*
	 * If the I/O is or was optional and therefore has no data, we need to
	 * simply discard it. We need to drop the vdev queue's lock to avoid a
	 * deadlock that we could encounter since this I/O will complete
	 * immediately.
	 */
	if (zio->io_flags & ZIO_FLAG_NODATA) {
		mutex_exit(&vq->vq_lock);
		zio_vdev_io_bypass(zio);
		zio_execute(zio);
		mutex_enter(&vq->vq_lock);
		goto again;
	}

	vdev_queue_pending_add(vq, zio);
	vq->vq_last_offset = zio->io_offset;

	return (zio);
}

zio_t *
vdev_queue_io(zio_t *zio)
{
	vdev_queue_t *vq = &zio->io_vd->vdev_queue;
	zio_t *nio;

	if (zio->io_flags & ZIO_FLAG_DONT_QUEUE)
		return (zio);

	/*
	 * Children i/os inherent their parent's priority, which might
	 * not match the child's i/o type.  Fix it up here.
	 */
	if (zio->io_type == ZIO_TYPE_READ) {
		if (zio->io_priority != ZIO_PRIORITY_SYNC_READ &&
		    zio->io_priority != ZIO_PRIORITY_ASYNC_READ &&
		    zio->io_priority != ZIO_PRIORITY_SCRUB)
			zio->io_priority = ZIO_PRIORITY_ASYNC_READ;
	} else if (zio->io_type == ZIO_TYPE_WRITE) {
		if (zio->io_priority != ZIO_PRIORITY_SYNC_WRITE &&
		    zio->io_priority != ZIO_PRIORITY_ASYNC_WRITE)
			zio->io_priority = ZIO_PRIORITY_ASYNC_WRITE;
	} else {
		ASSERT(zio->io_type == ZIO_TYPE_FREE);
		zio->io_priority = ZIO_PRIORITY_TRIM;
	}

	zio->io_flags |= ZIO_FLAG_DONT_CACHE | ZIO_FLAG_DONT_QUEUE;

	mutex_enter(&vq->vq_lock);
	zio->io_timestamp = gethrtime();
	vdev_queue_io_add(vq, zio);
	nio = vdev_queue_io_to_issue(vq);
	mutex_exit(&vq->vq_lock);

	if (nio == NULL)
		return (NULL);

	if (nio->io_done == vdev_queue_agg_io_done) {
		zio_nowait(nio);
		return (NULL);
	}

	return (nio);
}

void
vdev_queue_io_done(zio_t *zio)
{
	vdev_queue_t *vq = &zio->io_vd->vdev_queue;
	zio_t *nio;

	mutex_enter(&vq->vq_lock);

	vdev_queue_pending_remove(vq, zio);

	vq->vq_io_complete_ts = gethrtime();

	while ((nio = vdev_queue_io_to_issue(vq)) != NULL) {
		mutex_exit(&vq->vq_lock);
		if (nio->io_done == vdev_queue_agg_io_done) {
			zio_nowait(nio);
		} else {
			zio_vdev_io_reissue(nio);
			zio_execute(nio);
		}
		mutex_enter(&vq->vq_lock);
	}

	mutex_exit(&vq->vq_lock);
}

/*
 * As these three methods are only used for load calculations we're not concerned
 * if we get an incorrect value on 32bit platforms due to lack of vq_lock mutex
 * use here, instead we prefer to keep it lock free for performance.
 */ 
int
vdev_queue_length(vdev_t *vd)
{
	return (avl_numnodes(&vd->vdev_queue.vq_active_tree));
}

uint64_t
vdev_queue_lastoffset(vdev_t *vd)
{
	return (vd->vdev_queue.vq_lastoffset);
}

void
vdev_queue_register_lastoffset(vdev_t *vd, zio_t *zio)
{
	vd->vdev_queue.vq_lastoffset = zio->io_offset + zio->io_size;
}